ADIPOQ

UniProt ID: Q15848
Organism: Homo sapiens
Review Status: COMPLETE
πŸ“ Provide Detailed Feedback

Gene Description

Adiponectin is a 244-residue, adipocyte-specific secreted protein and one of the most abundant hormones in human plasma, circulating at roughly 5-30 ug/ml. After removal of an 18-residue signal peptide the mature chain comprises a short variable N-terminal segment, a collagen-like domain of 22 Gly-X-Y repeats (residues 42-107) and a C-terminal globular C1q domain (108-244), the architecture that places it in the C1q/TNF-related (CTRP) superfamily alongside the complement C1q chains. Three chains associate through a triple helix of their collagen-like domains and hydrophobic contacts within the globular heads to form the obligate trimer; trimers then join through interchain disulfide bonds into hexamers and 12- to 18-mer high-molecular-weight complexes. Extensive post-translational modification of the collagen domain - proline and lysine hydroxylation, glucosyl-galactosyl glycosylation of the hydroxylysines, and sialylation of O-linked glycans on N-terminal threonines - governs how much high-molecular-weight complex is assembled and secreted and how long the protein persists in blood. Oligomeric state determines which signal adiponectin delivers. Trimers preferentially activate AMP-activated protein kinase in skeletal muscle, whereas hexameric and high-molecular-weight species activate NF-kB and are the only forms that bind T-cadherin. Signalling runs through two seven-transmembrane receptors, AdipoR1 and AdipoR2, which are unrelated to G-protein-coupled receptors and recruit the adaptor APPL1, and through the GPI-anchored receptor T-cadherin. Downstream, AMPK and PPAR-alpha activation increase fatty-acid oxidation and glucose uptake in muscle and suppress gluconeogenic gene expression in liver, which is the basis of adiponectin's insulin-sensitising and glucose-lowering action; infusion reduces endogenous glucose production by around two thirds. Beyond metabolism, adiponectin acts on the vessel wall and on myeloid cells. It binds PDGF-BB directly and sequesters it from its receptor, suppressing smooth-muscle proliferation and migration; it blocks TNF-induced NF-kB activation and adhesion-molecule expression in endothelium through a cAMP/PKA-dependent route; it lowers scavenger-receptor-mediated lipid uptake and raises ABCA1-dependent cholesterol efflux in macrophages; and it restrains myelomonocytic proliferation and phagocytosis. These effects are not uniformly anti-inflammatory - high-molecular-weight adiponectin induces MCP-1 and IL-8 in peripheral blood mononuclear cells and microvascular endothelium while the low-molecular-weight form does not - so the sign of the response depends on multimer state and on the target cell. Plasma adiponectin falls with visceral obesity and insulin resistance, and rare variants that impair high-molecular-weight assembly or secretion cause autosomal dominant adiponectin deficiency.

Proposed New Ontology Terms

regulation of glomerular permeability to albumin

Definition: Any process that modulates the rate, frequency or extent at which albumin crosses the glomerular filtration barrier from blood into the urinary space.

Justification: GO can currently express changes in urinary albumin only as renal albumin ABSORPTION (GO:0097018 and its regulation terms GO:2000532-2000534), which sits under GO:0097017 renal protein absorption - uptake from the nephron lumen. That is a different process from permselectivity of the glomerular filtration barrier, and the mismatch has produced a real mis-annotation: PMID:18431508 shows adiponectin acting through AMPK to reduce podocyte permeability to albumin, with no measurement of tubular reabsorption anywhere in the paper, yet both human ADIPOQ and mouse Adipoq carry GO:2000534 positive regulation of renal albumin absorption. Podocyte-barrier biology is a large and active field (nephrin, podocin, slit diaphragm) that has no adequate GO process term for its central readout.

Parent term: regulation of glomerular filtration

Supporting Evidence:

regulation of sympathetic nervous system activity

Definition: Any process that modulates the frequency, rate or extent of efferent signalling by the sympathetic division of the autonomic nervous system to its target organs.

Justification: Searching GO for 'sympathetic' returns only developmental terms (GO:0048485 sympathetic nervous system development, GO:0061549, GO:0097490, GO:0097491, GO:0097492, GO:1903045) and for 'autonomic' only GO:0048483 and GO:1901166. There is no term for modulating sympathetic outflow, which is a standard, directly measurable physiological readout. In its absence, PMID:17327472's measurement of reduced renal sympathetic nerve activity was annotated to GO:0050805 negative regulation of synaptic transmission, a term about communication across an individual synapse, which the study did not assay.

Parent term: regulation of system process

Supporting Evidence:

Existing Annotations Review

GO Term Evidence Action Reason
GO:0005576 extracellular region
IBA
GO_REF:0000033
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree. The IBA sits on node PTN008355511, whose human reach is ADIPOQ, C1QA, C1QB, C1QC, C1QTNF2, C1QTNF5, C1QTNF7, C1QTNF9 and C1QTNF9B. For a clade mixing the complement C1q chains with the CTRP adipokines the generic extracellular term is the correct least common ancestor - refining it would mean picking one donor's compartment over another's. All 15 donor tokens resolve and all 15 carry their own experimental evidence for the term, and the WITH/FROM includes UniProtKB:Q15848 itself.
Propagation Review
Root cause: NO FAILURE CORE
Sources checked:
PANTHER:PTN008355511
C1q/CTRP family node; 9 human genes reached
UniProtKB:Q15848
self-reference - target is its own IBD seed
UniProtKB:P02745
human C1QA, own IDA
UniProtKB:P02746
human C1QB, own IDA
UniProtKB:P02747
human C1QC, own IDA
UniProtKB:Q3Y5Z3
bovine ADIPOQ, own IDA
UniProtKB:Q9BXJ0
human C1QTNF5, own IDA
MGI:MGI:106675
mouse Adipoq (Q60994), own EXP/IDA
MGI:MGI:1916433
mouse C1qtnf2 (Q9D8U4), own IDA
MGI:MGI:1925911
mouse C1qtnf7 (Q8BVD7), own IDA
MGI:MGI:2385958
mouse C1qtnf5 (Q8K479), own IDA/HDA
MGI:MGI:3045252
mouse C1qtnf9 (Q4ZJN1), own IDA
RGD:1306716
rat C1qa (P31720), own EXP
RGD:1306828
rat C1qc (P31722), own EXP
RGD:2229
rat C1qb (P31721), own IDA
RGD:628748
rat Adipoq (F7FPS2, TrEMBL), own IDA
GO:0005179 hormone activity
IBA
GO_REF:0000033
ACCEPT
Summary: Adiponectin is a bona fide adipocyte-derived endocrine hormone and this is its core molecular function.
Reason: GO:0005179 is defined as the action of 'any substance formed in very small amounts in one specialized organ or group of cells and carried (sometimes in the bloodstream) to another organ or group of cells in the same organism, upon which it has a specific regulatory action'. Adiponectin is made only by adipocytes, circulates in plasma at 5-30 ug/ml, and acts on liver, skeletal muscle, macrophages, endothelium and kidney through AdipoR1, AdipoR2 and T-cadherin. The definition is met exactly, and this term - not the coarser signaling receptor binding, and not extracellular matrix or interaction-screen terms - is what the gene is for. The IBA is self-referential - its WITH/FROM is MGI:MGI:106675|PANTHER:PTN008559544|UniProtKB:Q15848, i.e. it includes ADIPOQ itself - so it records a PAINT curator judging this function core rather than importing it from elsewhere. Both donor tokens resolve and both carry their own experimental evidence for the term (MGI:MGI:106675 = Q60994 ADIPO_MOUSE, IDA). Node PTN008559544 reaches exactly one human gene, ADIPOQ, checked against all 82 human GO:0005179 IBA rows.
Propagation Review
Root cause: NO FAILURE CORE
Sources checked:
MGI:MGI:106675
mouse Adipoq (Q60994, Swiss-Prot); carries its own IDA for this term
PANTHER:PTN008559544
ancestral node; human reach is exactly ADIPOQ
UniProtKB:Q15848
self-reference - the target gene is its own IBD seed
GO:0010642 negative regulation of platelet-derived growth factor receptor signaling pathway
IBA
GO_REF:0000033
ACCEPT
Summary: Adiponectin binds PDGF-BB and blocks its access to the receptor; a directly demonstrated, mechanistically explicit process.
Reason: This is one of the few adiponectin activities with a directly measured molecular mechanism in human cells. PMID:12070119 showed that human adiponectin binds PDGF-BB itself and thereby blocks the ligand from reaching its receptor on human aortic smooth muscle cells, suppressing PDGF beta-receptor autophosphorylation. Ligand sequestration is a direct participation in the negative regulation of that pathway, not a downstream physiological consequence.
Propagation Review
Root cause: NO FAILURE CORE
Sources checked:
MGI:MGI:106675
mouse Adipoq (Q60994); carries its own IDA for this term
PANTHER:PTN008559544
human reach is exactly ADIPOQ; this is the only human GO:0010642 IBA row in GOA
UniProtKB:Q15848
self-reference
GO:0045599 negative regulation of fat cell differentiation
IBA
GO_REF:0000033
ACCEPT
Summary: Adiponectin restrains adipocyte differentiation in a paracrine negative-feedback loop.
Reason: Recombinant adiponectin blocks adipogenesis in bone-marrow stromal cultures through a COX-2/prostaglandin-dependent paracrine loop, establishing preadipocytes as direct adiponectin targets. Combined with the gene's identity as the adipocyte's own secreted product, this is a well-founded feedback function rather than an incidental phenotype.
Propagation Review
Root cause: NO FAILURE CORE
Sources checked:
MGI:MGI:106675
mouse Adipoq (Q60994)
PANTHER:PTN008559544
human reach is exactly ADIPOQ
UniProtKB:Q15848
self-reference
GO:0005576 extracellular region
IEA
GO_REF:0000044
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree.
GO:0005515 protein binding
IPI
PMID:25910212
Widespread macromolecular interaction perturbations in human...
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SGTA (O43765)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. This partner (SGTA) is the only one recorded in more than two of the seven IntAct publications, appearing in PMID:25910212, PMID:31515488 and PMID:32296183 - but all three are two-hybrid interactome datasets from the same methodological lineage, so the repetition is the same assay run again rather than orthogonal validation. SGTA is a cytosolic co-chaperone for mislocalised tail-anchored proteins, a well-known source of promiscuous hits. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:25910212
Widespread macromolecular interaction perturbations in human...
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (BNIP3L (O60238)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:25910212
Widespread macromolecular interaction perturbations in human...
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SYNE4 (Q8N205-2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:25910212
Widespread macromolecular interaction perturbations in human...
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (KASH5 (Q8N6L0)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:31515488
Extensive disruption of protein interactions by genetic vari...
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SGTA (O43765)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. This partner (SGTA) is the only one recorded in more than two of the seven IntAct publications, appearing in PMID:25910212, PMID:31515488 and PMID:32296183 - but all three are two-hybrid interactome datasets from the same methodological lineage, so the repetition is the same assay run again rather than orthogonal validation. SGTA is a cytosolic co-chaperone for mislocalised tail-anchored proteins, a well-known source of promiscuous hits. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SLC22A23 (A1A5C7-2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (PEX12 (O00623)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (NEMP1 (O14524-2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (TMPRSS2 (O15393-2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (GPR42 (O15529)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (TNFSF14 (O43557)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SGTA (O43765)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. This partner (SGTA) is the only one recorded in more than two of the seven IntAct publications, appearing in PMID:25910212, PMID:31515488 and PMID:32296183 - but all three are two-hybrid interactome datasets from the same methodological lineage, so the repetition is the same assay run again rather than orthogonal validation. SGTA is a cytosolic co-chaperone for mislocalised tail-anchored proteins, a well-known source of promiscuous hits. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (COQ9 (O75208)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (KCNK5 (O95279)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CLDN9 (O95484)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (ASGR2 (P07307-3)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CD79A (P11912)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (FCGR1A (P12314)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (PVR (P15151)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (MUC1 (P15941-11)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (KEL (P23276)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (ERP29 (P30040)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CNR2 (P34972)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (IFNGR2 (P38484)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CAMLG (P49069)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (FASN (P49327)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (BIK (Q13323)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (AQP6 (Q13520)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (TMEM35A (Q53FP2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (FAM209A (Q5JX71)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SLC71A2 (Q5SR56)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (MRM1 (Q6IN84)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (BTNL9 (Q6UXG8-3)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (MFSD6 (Q6ZSS7)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SLC35E3 (Q7Z769)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CPLX4 (Q7Z7G2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (TMEM179B (Q7Z7N9)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (AMIGO1 (Q86WK6)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CLEC10A (Q8IUN9)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SYNE4 (Q8N205-2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (HSD17B11 (Q8NBQ5)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (PROKR2 (Q8NFJ6)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (FNDC9 (Q8TBE3)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CD200R1 (Q8TD46-4)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (GPR152 (Q8TDT2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CCDC107 (Q8WV48)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (BSND (Q8WZ55)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SYAP1 (Q96A49)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CREB3L1 (Q96BA8)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SGTB (Q96EQ0)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (MS4A3 (Q96HJ5)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SIGLEC12 (Q96PQ1)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (TMEM237 (Q96Q45-2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CYBC1 (Q9BQA9)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (BCL2L13 (Q9BXK5)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (MFF (Q9GZY8-5)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SLC35H1 (Q9NQQ7-3)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (SCN3B (Q9NY72)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (CLEC2D (Q9UHP7-3)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (TRIM35 (Q9UPQ4-2)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32296183
A reference map of the human binary protein interactome.
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (FKBP7 (Q9Y680)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005515 protein binding
IPI
PMID:32814053
Interactome Mapping Provides a Network of Neurodegenerative ...
MARK AS OVER ANNOTATED
Summary: Unreplicated interactome-screen hit (HTT (P42858)); no orthogonal evidence and no plausible compartment in which secreted adiponectin could meet it.
Reason: This row comes from a systematic binary-interactome screen, not from a hypothesis-led experiment on adiponectin. IntAct records 286 interactions for Q15848 across only 7 publications, of which 171 are from the single HuRI dataset (PMID:32296183) and 196 are two-hybrid, logged under five sub-method names (two hybrid array, validated two hybrid, two hybrid prey pooling approach, two hybrid pooling, two hybrid bait and prey pooling approach) - which is why UniProt shows NbExp=3 on nearly every partner even though there is one experiment. 240 of 264 IntAct partners appear in exactly one publication, and no orthogonal biophysical assay (SPR, ITC, co-immunoprecipitation of endogenous protein) supports any of them. Adiponectin is synthesised with a cleaved signal peptide (SIGNAL 1..18) and its mature chain is secreted; two-hybrid requires both partners to reconstitute a transcription factor in the yeast nucleus, a compartment the native protein never enters. Its assembly into disulfide-linked hexamers and HMW multimers, and the hydroxylation/glycosylation those multimers depend on, also cannot occur in that assay. This partner (HTT, huntingtin) is a 3142-residue cytosolic scaffold and one of the most frequently reported hubs in interactome datasets; the reference here is a neurodegeneration-focused interactome (PMID:32814053, 20010 annotations in GOA). Marked as over-annotated rather than removed: an unreplicated screen hit is unmeasured, not refuted, and a bare protein binding term carries no functional information either way.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
There is **no orthogonal biophysical assay** anywhere in the set
GO:0005179 hormone activity
IEA
GO_REF:0000107
ACCEPT
Summary: Adiponectin is a bona fide adipocyte-derived endocrine hormone and this is its core molecular function.
Reason: GO:0005179 is defined as the action of 'any substance formed in very small amounts in one specialized organ or group of cells and carried (sometimes in the bloodstream) to another organ or group of cells in the same organism, upon which it has a specific regulatory action'. Adiponectin is made only by adipocytes, circulates in plasma at 5-30 ug/ml, and acts on liver, skeletal muscle, macrophages, endothelium and kidney through AdipoR1, AdipoR2 and T-cadherin. The definition is met exactly, and this term - not the coarser signaling receptor binding, and not extracellular matrix or interaction-screen terms - is what the gene is for.
GO:0010642 negative regulation of platelet-derived growth factor receptor signaling pathway
IEA
GO_REF:0000107
ACCEPT
Summary: Adiponectin binds PDGF-BB and blocks its access to the receptor; a directly demonstrated, mechanistically explicit process.
Reason: This is one of the few adiponectin activities with a directly measured molecular mechanism in human cells. PMID:12070119 showed that human adiponectin binds PDGF-BB itself and thereby blocks the ligand from reaching its receptor on human aortic smooth muscle cells, suppressing PDGF beta-receptor autophosphorylation. Ligand sequestration is a direct participation in the negative regulation of that pathway, not a downstream physiological consequence.
GO:0030336 negative regulation of cell migration
IEA
GO_REF:0000107
KEEP AS NON CORE
Summary: Adiponectin restrains migration of vascular smooth muscle and mesangial cells.
Reason: Transferred by ISS from mouse Adipoq, whose GO:0030336 IDA comes from PMID:19460854 on PDGF-induced mesangial cell migration. Mechanistically this is the same PDGF-BB sequestration captured more precisely by GO:0010642 and GO:1904753; kept as the general vascular-remodelling statement.
GO:0038002 endocrine signaling
IEA
GO_REF:0000107
ACCEPT
Summary: Adiponectin is carried in the bloodstream from adipocytes to distant target organs; this is the process form of its core hormone function.
Reason: GO:0038002 is defined as signalling 'where an endocrine hormone is transported from the signal-producing cell to the receiving cell via the circulatory system (via blood, lymph or cerebrospinal fluid). The signaling cell and the receiving cell are often distant to each other.' Adiponectin is made only by adipocytes, circulates at 5-30 ug/ml and acts on liver, muscle, endothelium, macrophages and kidney. The definition is met exactly. Although the row is an Ensembl Compara projection, the conclusion is established by the human literature independently of the projection, so it is accepted as core rather than downgraded on evidence code alone.
GO:0042802 identical protein binding
IEA
GO_REF:0000107
ACCEPT
Summary: Adiponectin self-associates into trimers, hexamers and HMW multimers; identical protein binding is correct and is the right grain for a homomultimer of variable stoichiometry.
Reason: UniProt SUBUNIT: 'Homomultimer. Forms trimers, hexamers and 12- to 18-mers.' The fundamental unit is a trimer built from a triple helix of the collagen-like domains plus hydrophobic contacts in the globular C1q domain; hexamers and HMW species follow from interchain disulfides at Cys-36 (mature Cys-22 numbering in PMID:14522956). GO:0042802 'Binding to an identical protein or proteins' is stoichiometry-neutral and therefore correct for all three forms.
GO:0045599 negative regulation of fat cell differentiation
IEA
GO_REF:0000107
ACCEPT
Summary: Adiponectin restrains adipocyte differentiation in a paracrine negative-feedback loop.
Reason: Recombinant adiponectin blocks adipogenesis in bone-marrow stromal cultures through a COX-2/prostaglandin-dependent paracrine loop, establishing preadipocytes as direct adiponectin targets. Combined with the gene's identity as the adipocyte's own secreted product, this is a well-founded feedback function rather than an incidental phenotype.
GO:0050728 negative regulation of inflammatory response
IEA
GO_REF:0000107
KEEP AS NON CORE
Summary: Adiponectin is broadly anti-inflammatory.
Reason: A general statement supported by the endothelial adhesion molecule, TNF and macrophage results. Kept as the umbrella term for a set of downstream immune effects, none of which is the gene's molecular function.
GO:0050996 positive regulation of lipid catabolic process
IEA
GO_REF:0000107
KEEP AS NON CORE
Summary: Adiponectin increases lipid catabolism in muscle and liver.
Reason: The organism-level expression of the AMPK-driven fatty-acid oxidation effect. Retained as the general statement; the specific, direction-bearing term proposed for the GO:0006635/GO:0019395 rows is GO:0046321.
GO:0071466 cellular response to xenobiotic stimulus
IEA
GO_REF:0000107
MARK AS OVER ANNOTATED
Summary: Compara projection of the same rosiglitazone-induction datum; same role inversion as the ISS row.
Reason: Ensembl Compara projection from mouse Adipoq, whose GO:0071466 IDA rests on PMID:19109165 showing that rosiglitazone induces adiponectin expression. The gene product is the output of the response, not a component of it.
Propagation Review
Root cause: SOURCE BAD
Failure modes: ROLE CONFLATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq GO:0071466 IDA from PMID:19109165
ensembl:ENSMUSP00000023593
Compara protein id for mouse Adipoq, as emitted in the GOA WITH/FROM
GO:0120162 positive regulation of cold-induced thermogenesis
IEA
GO_REF:0000107
KEEP AS NON CORE
Summary: Ensembl Compara projection of the mouse thermogenesis rows; retained at the same non-core grade as the ISS rows it mirrors.
Reason: GOA cites GO:0120162 (positive) and GO:0120163 (negative) regulation of cold-induced thermogenesis to the SAME two references, PMID:24531262 and PMID:26166748 - a full 2x2 cross-product. The same evidence cannot support a proposition and its negation. Reading the two papers assigns each to exactly one direction: PMID:24531262 (Diabetologia 2014) reports that Adipoq-knockout mice run HOTTER and express MORE UCP1, i.e. adiponectin SUPPRESSES thermogenesis; PMID:26166748 (Cell Metab 2015) reports that the cold-induced thermogenic program is IMPAIRED in adiponectin-knockout mice, i.e. adiponectin PROMOTES it. The same cross-product is present on mouse Adipoq (Q60994) as four IMP rows from the same two references, so the defect originates upstream at MGI and reaches human through these ISS transfers. The IEA rows are Compara projections from mouse Adipoq and inherit the same evidence. They are kept because each direction does have one correctly-cited paper behind it; the fix belongs upstream, where the four mouse IMP rows should become two.
GO:0120163 negative regulation of cold-induced thermogenesis
IEA
GO_REF:0000107
KEEP AS NON CORE
Summary: Ensembl Compara projection of the mouse thermogenesis rows; retained at the same non-core grade as the ISS rows it mirrors.
Reason: GOA cites GO:0120162 (positive) and GO:0120163 (negative) regulation of cold-induced thermogenesis to the SAME two references, PMID:24531262 and PMID:26166748 - a full 2x2 cross-product. The same evidence cannot support a proposition and its negation. Reading the two papers assigns each to exactly one direction: PMID:24531262 (Diabetologia 2014) reports that Adipoq-knockout mice run HOTTER and express MORE UCP1, i.e. adiponectin SUPPRESSES thermogenesis; PMID:26166748 (Cell Metab 2015) reports that the cold-induced thermogenic program is IMPAIRED in adiponectin-knockout mice, i.e. adiponectin PROMOTES it. The same cross-product is present on mouse Adipoq (Q60994) as four IMP rows from the same two references, so the defect originates upstream at MGI and reaches human through these ISS transfers. The IEA rows are Compara projections from mouse Adipoq and inherit the same evidence. They are kept because each direction does have one correctly-cited paper behind it; the fix belongs upstream, where the four mouse IMP rows should become two.
GO:2000534 positive regulation of renal albumin absorption
IEA
GO_REF:0000107
MARK AS OVER ANNOTATED
Summary: Compara projection of the same mouse podocyte result; inherits the same process mismatch.
Reason: GO:2000534 sits under GO:0097017 renal protein absorption, defined as proteins 'taken up from the collecting ducts, glomerulus and proximal and distal loops of the nephron'. PMID:18431508 measures something else entirely: it shows that adiponectin-knockout mice have podocyte foot-process effacement and increased albuminuria, and that adiponectin acting through AMPK reduces podocyte PERMEABILITY to albumin. The full text contains no measurement of tubular reabsorption; the words absorption, reabsorption and tubular do not appear in the extracted text at all. So adiponectin restores the glomerular filtration barrier - less albumin crosses it - rather than increasing uptake downstream. GO has no term for regulation of glomerular permeability to albumin, which is very likely why this one was chosen; that gap is filed under proposed_new_terms. Marked over-annotated rather than removed because the phenotype (less urinary albumin) is real and the term is the closest existing fit.
Propagation Review
Root cause: TERM SCOPING PROBLEM
Failure modes: ROLE CONFLATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq GO:2000534 IMP from PMID:18431508
ensembl:ENSMUSP00000023593
Compara protein id for mouse Adipoq, as emitted in the GOA WITH/FROM
GO:2000584 negative regulation of platelet-derived growth factor receptor-alpha signaling pathway
IEA
GO_REF:0000107
KEEP AS NON CORE
Summary: Adiponectin dampens PDGF-alpha-receptor signalling downstream of the receptor.
Reason: Worth stating precisely, because PMID:12070119 reports that adiponectin reduced PDGF-AA-stimulated ERK phosphorylation 'without affecting autophosphorylation of PDGF alpha-receptor', and that adiponectin does not bind PDGF-AA. So the alpha-receptor pathway is inhibited at a post-receptor step, unlike the beta-receptor pathway where ligand sequestration acts upstream. The term is about the pathway, which includes downstream steps, so it stands - but it should not be read as ligand sequestration.
GO:2000590 negative regulation of metanephric mesenchymal cell migration
IEA
GO_REF:0000107
KEEP AS NON CORE
Summary: Adiponectin inhibits PDGF-driven migration of mesangial cells.
Reason: Transferred from mouse Adipoq, whose annotation rests on PMID:19460854, 'Inhibitory effects of adiponectin on platelet-derived growth factor-induced mesangial cell migration'. The term names metanephric MESENCHYMAL cells, the embryonic progenitors, whereas the assay used cultured mesangial cells; GO has no mesangial cell migration term, so this is the closest available fit rather than an error. Mechanistically it is again PDGF antagonism.
GO:0140149 non-collagenous component of interstitial matrix
TAS
PMID:36399478
MatrisomeDB 2.0: 2023 updates to the ECM-protein knowledge d...
MARK AS OVER ANNOTATED
Summary: A MatrisomeDB database import: one reference gives this term to 272 entities. The term's definition is about structural matrix glycoproteins, which adiponectin is not.
Reason: PMID:36399478 is the MatrisomeDB 2.0 database-update paper. The reference-projection test returns 285 annotations over 274 distinct entities, of which GO:0140149 alone reaches 272, all assigned by GO_Central - one reference giving 272 entities one identical cellular-component term is a bulk import, not 272 traceable author statements. GO:0140149 is defined as 'the non-collagenous component of interstitial extracellular matrices, including glycoprotein like fibronectin and elastin', i.e. the structural non-collagenous compartment. Adiponectin is a 30 kDa circulating hormone that appears in matrisome fractions because it adsorbs to tissue, not because it builds interstitial matrix. Marked over-annotated rather than removed because its presence in ECM preparations is a real, reproducible observation and is already recorded by the GO:0031012 HDA row at the grain the data supports.
Propagation Review
Root cause: PROPAGATION BAD
Failure modes: ROLE CONFLATION
Sources checked:
PMID:36399478
MatrisomeDB 2.0; 272 entities receive GO:0140149 from this single reference
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
`GO:0140149` from **PMID:36399478 is a database import**
GO:0005179 hormone activity
ISS
GO_REF:0000024
ACCEPT
Summary: Adiponectin is a bona fide adipocyte-derived endocrine hormone and this is its core molecular function.
Reason: GO:0005179 is defined as the action of 'any substance formed in very small amounts in one specialized organ or group of cells and carried (sometimes in the bloodstream) to another organ or group of cells in the same organism, upon which it has a specific regulatory action'. Adiponectin is made only by adipocytes, circulates in plasma at 5-30 ug/ml, and acts on liver, skeletal muscle, macrophages, endothelium and kidney through AdipoR1, AdipoR2 and T-cadherin. The definition is met exactly, and this term - not the coarser signaling receptor binding, and not extracellular matrix or interaction-screen terms - is what the gene is for.
GO:0033211 adiponectin-activated signaling pathway
ISS
GO_REF:0000024
ACCEPT
Summary: The adiponectin-activated signalling pathway is, by definition, this gene's core process.
Reason: GO:0033211 is defined as 'the series of molecular signals initiated by adiponectin binding to its receptor on the surface of a cell, and ending with the regulation of a downstream cellular process'. This is the gene's own pathway and its core biological process.
GO:0050996 positive regulation of lipid catabolic process
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Adiponectin increases lipid catabolism in muscle and liver.
Reason: The organism-level expression of the AMPK-driven fatty-acid oxidation effect. Retained as the general statement; the specific, direction-bearing term proposed for the GO:0006635/GO:0019395 rows is GO:0046321.
GO:0005576 extracellular region
EXP
PMID:8947845
Isolation and characterization of GBP28, a novel gelatin-bin...
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree.
Supporting Evidence:
PMID:8947845
The clone encodes a polypeptide of 244 amino acids with a secretory signal sequence at the amino terminus
GO:0005515 protein binding
IPI
PMID:16622416
APPL1 binds to adiponectin receptors and mediates adiponecti...
MODIFY
Summary: The partner is adiponectin's own receptor AdipoR1; signaling receptor binding is the informative term.
Reason: ADIPOR1 is one of the two seven-transmembrane adiponectin receptors cloned by expression cloning in PMID:12802337, and the reciprocal row exists on ADIPOR1 (Q96A54 carries GO:0005515 IPI PMID:16622416 with UniProtKB:Q15848). Replacing bare protein binding with GO:0005102 signaling receptor binding records what the partnership is. One caveat worth stating: PMID:16622416 is primarily an APPL1 paper, and what it demonstrates about adiponectin is that 'APPL1 interacts with adiponectin receptors in mammalian cells and the interaction is stimulated by adiponectin' - a ligand-dependent effect on a receptor complex rather than a direct binding measurement. The direct adiponectin-AdipoR1 interaction is established by PMID:12802337 instead.
Proposed replacements: signaling receptor binding
Supporting Evidence:
PMID:16622416
APPL1 interacts with adiponectin receptors in mammalian cells and the interaction is stimulated by adiponectin
PMID:12802337
they serve as receptors for globular and full-length adiponectin
GO:0005179 hormone activity
IDA
PMID:11222466
Adipocyte-derived plasma protein, adiponectin, suppresses li...
ACCEPT
Summary: Adiponectin is a bona fide adipocyte-derived endocrine hormone and this is its core molecular function.
Reason: GO:0005179 is defined as the action of 'any substance formed in very small amounts in one specialized organ or group of cells and carried (sometimes in the bloodstream) to another organ or group of cells in the same organism, upon which it has a specific regulatory action'. Adiponectin is made only by adipocytes, circulates in plasma at 5-30 ug/ml, and acts on liver, skeletal muscle, macrophages, endothelium and kidney through AdipoR1, AdipoR2 and T-cadherin. The definition is met exactly, and this term - not the coarser signaling receptor binding, and not extracellular matrix or interaction-screen terms - is what the gene is for.
Supporting Evidence:
PMID:11222466
The adipocyte-derived plasma protein adiponectin suppressed macrophage-to-foam cell transformation
GO:0005179 hormone activity
IDA
PMID:18703020
Adiponectin prevents atherosclerosis by increasing cholester...
ACCEPT
Summary: Adiponectin is a bona fide adipocyte-derived endocrine hormone and this is its core molecular function.
Reason: GO:0005179 is defined as the action of 'any substance formed in very small amounts in one specialized organ or group of cells and carried (sometimes in the bloodstream) to another organ or group of cells in the same organism, upon which it has a specific regulatory action'. Adiponectin is made only by adipocytes, circulates in plasma at 5-30 ug/ml, and acts on liver, skeletal muscle, macrophages, endothelium and kidney through AdipoR1, AdipoR2 and T-cadherin. The definition is met exactly, and this term - not the coarser signaling receptor binding, and not extracellular matrix or interaction-screen terms - is what the gene is for.
Supporting Evidence:
PMID:18703020
APN up-regulated the expression of ABCA1 in human macrophages
GO:0005576 extracellular region
IDA
PMID:11222466
Adipocyte-derived plasma protein, adiponectin, suppresses li...
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree.
GO:0033211 adiponectin-activated signaling pathway
IDA
PMID:18703020
Adiponectin prevents atherosclerosis by increasing cholester...
ACCEPT
Summary: The adiponectin-activated signalling pathway is, by definition, this gene's core process.
Reason: GO:0033211 is defined as 'the series of molecular signals initiated by adiponectin binding to its receptor on the surface of a cell, and ending with the regulation of a downstream cellular process'. This is the gene's own pathway and its core biological process.
GO:0048261 negative regulation of receptor-mediated endocytosis
IDA
PMID:11222466
Adipocyte-derived plasma protein, adiponectin, suppresses li...
KEEP AS NON CORE
Summary: Adiponectin reduces scavenger-receptor-mediated uptake in macrophages.
Reason: The same class A scavenger receptor result from PMID:11222466, expressed as an endocytosis term. Downstream.
GO:0060621 negative regulation of cholesterol import
IDA
PMID:11222466
Adipocyte-derived plasma protein, adiponectin, suppresses li...
KEEP AS NON CORE
Summary: Adiponectin reduces cholesterol uptake by macrophages.
Reason: The uptake half of the PMID:11222466 foam-cell result, complementary to the GO:0010875 efflux row. Downstream.
GO:0141163 positive regulation of cAMP/PKA signal transduction
IDA
PMID:10982546
Adiponectin, an adipocyte-derived plasma protein, inhibits e...
KEEP AS NON CORE
Summary: Adiponectin raises cAMP and activates PKA in endothelial cells.
Reason: PMID:10982546 attributed adiponectin's NF-kB suppression to a cAMP-PKA-dependent pathway. This is the proximal second-messenger step of one branch of adiponectin signalling; it is kept non-core because the receptor-proximal event is already covered by GO:0033211.
GO:0005201 extracellular matrix structural constituent
RCA
PMID:28675934
Characterization of the Extracellular Matrix of Normal and D...
REMOVE
Summary: Adiponectin is a circulating hormone, not a structural component of the extracellular matrix; this row is a bulk computational block that converted matrisome detection into a structural role.
Reason: Three independent grounds. (1) Provenance: PMID:28675934 is an ECM proteomics survey, and in GOA it assigns GO:0005201 by RCA to 41 entities, together with GO:0030020, GO:0030021 and GO:0030023 - the collagen tensile-strength terms - to 31, 8 and 4 more. That is a computational block over a matrisome protein list, not a measurement of adiponectin's contribution to matrix integrity. (2) The term's definition is 'the action of a molecule that contributes to the structural integrity of the extracellular matrix'; no such contribution has been reported for adiponectin, and UniProt places the mature protein in 'Secreted' with no matrix role. (3) The likely trigger is the collagen-like domain (residues 42-107), but UniProt's SUBUNIT is explicit that those repeats build adiponectin's own homotrimer: the low-molecular-weight trimers are assembled via non-covalent interactions of the collagen-like domains in a triple helix. They do not crosslink a matrix. This is a fold-to-function error arriving through a bulk RCA import rather than through the retired Swiss-Prot-keyword route. Adiponectin does adsorb to matrix in injured tissue, but that is binding, not being a structural constituent, and no GO row currently makes the binding claim.
Propagation Review
Root cause: PROPAGATION BAD
Failure modes: ROLE CONFLATION GRANULARITY MISMATCH
Sources checked:
PMID:28675934
bulk RCA block: 41 entities receive GO:0005201 from this one ECM-proteomics reference
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-bioinformatics/RESULTS.md
`GO:0005201` from **PMID:28675934 is a bulk `RCA` block**
GO:0031012 extracellular matrix
HDA
PMID:28675934
Characterization of the Extracellular Matrix of Normal and D...
KEEP AS NON CORE
Summary: Adiponectin is reproducibly detected in tissue ECM preparations, which reflects adsorption of a circulating protein rather than matrix residency as its site of action.
Reason: This is a high-throughput proteomics detection: PMID:28675934 assigns GO:0031012 by HDA to 135 entities. Adiponectin genuinely accumulates in injured vessel wall and on hepatocyte matrix, so the localisation is not spurious. It is non-core because adiponectin's function is executed on receptors at the cell surface, and because being recovered in a matrisome fraction is a much weaker statement than the structural claim GO:0005201 makes from the same reference.
GO:0120162 positive regulation of cold-induced thermogenesis
ISS
PMID:24531262
Adiponectin reduces thermogenesis by inhibiting brown adipos...
REMOVE
Summary: Sign inversion: this reference reports that adiponectin SUPPRESSES thermogenesis, so it cannot support the positive-regulation term.
Reason: GOA cites GO:0120162 (positive) and GO:0120163 (negative) regulation of cold-induced thermogenesis to the SAME two references, PMID:24531262 and PMID:26166748 - a full 2x2 cross-product. The same evidence cannot support a proposition and its negation. Reading the two papers assigns each to exactly one direction: PMID:24531262 (Diabetologia 2014) reports that Adipoq-knockout mice run HOTTER and express MORE UCP1, i.e. adiponectin SUPPRESSES thermogenesis; PMID:26166748 (Cell Metab 2015) reports that the cold-induced thermogenic program is IMPAIRED in adiponectin-knockout mice, i.e. adiponectin PROMOTES it. The same cross-product is present on mouse Adipoq (Q60994) as four IMP rows from the same two references, so the defect originates upstream at MGI and reaches human through these ISS transfers. This particular row pairs the POSITIVE term with the paper that demonstrates suppression, and is therefore removable on the reference's own conclusion without adjudicating the underlying biological disagreement. The same paper correctly supports the GO:0120163 row, which is kept.
Propagation Review
Root cause: SOURCE BAD
Failure modes: REGULATORY SIGN INVERSION SOURCE MISCITATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq carries the identical cross-product: GO:0120162 and GO:0120163 each by IMP from BOTH PMID:24531262 and PMID:26166748
Supporting Evidence:
PMID:24531262
This study demonstrates that adiponectin suppresses thermogenesis
GO:0120162 positive regulation of cold-induced thermogenesis
ISS
PMID:26166748
Adiponectin Enhances Cold-Induced Browning of Subcutaneous A...
KEEP AS NON CORE
Summary: Correctly paired half of the cross-product: this reference does show adiponectin promoting cold-induced thermogenesis.
Reason: GOA cites GO:0120162 (positive) and GO:0120163 (negative) regulation of cold-induced thermogenesis to the SAME two references, PMID:24531262 and PMID:26166748 - a full 2x2 cross-product. The same evidence cannot support a proposition and its negation. Reading the two papers assigns each to exactly one direction: PMID:24531262 (Diabetologia 2014) reports that Adipoq-knockout mice run HOTTER and express MORE UCP1, i.e. adiponectin SUPPRESSES thermogenesis; PMID:26166748 (Cell Metab 2015) reports that the cold-induced thermogenic program is IMPAIRED in adiponectin-knockout mice, i.e. adiponectin PROMOTES it. The same cross-product is present on mouse Adipoq (Q60994) as four IMP rows from the same two references, so the defect originates upstream at MGI and reaches human through these ISS transfers. This row is the correctly paired one for the positive term and is retained. It is non-core because it is a whole-animal knockout phenotype several steps downstream of receptor engagement, and because the field has not reconciled it with the opposite result.
Supporting Evidence:
PMID:26166748
Chronic cold exposure-induced accumulation of M2 macrophages, activation of beige cells, and thermogenic program were markedly impaired in scWAT of adiponectin knockout (ADN KO) mice
GO:0120163 negative regulation of cold-induced thermogenesis
ISS
PMID:24531262
Adiponectin reduces thermogenesis by inhibiting brown adipos...
KEEP AS NON CORE
Summary: Correctly paired half of the cross-product: this reference does show adiponectin suppressing cold-induced thermogenesis.
Reason: GOA cites GO:0120162 (positive) and GO:0120163 (negative) regulation of cold-induced thermogenesis to the SAME two references, PMID:24531262 and PMID:26166748 - a full 2x2 cross-product. The same evidence cannot support a proposition and its negation. Reading the two papers assigns each to exactly one direction: PMID:24531262 (Diabetologia 2014) reports that Adipoq-knockout mice run HOTTER and express MORE UCP1, i.e. adiponectin SUPPRESSES thermogenesis; PMID:26166748 (Cell Metab 2015) reports that the cold-induced thermogenic program is IMPAIRED in adiponectin-knockout mice, i.e. adiponectin PROMOTES it. The same cross-product is present on mouse Adipoq (Q60994) as four IMP rows from the same two references, so the defect originates upstream at MGI and reaches human through these ISS transfers. This row is the correctly paired one for the negative term and is retained, non-core for the same reasons as its positive counterpart.
Supporting Evidence:
PMID:24531262
The CBTs of adiponectin knockout mice (Adipoq(-/-)) were significantly higher than those of wild type (WT) mice
GO:0120163 negative regulation of cold-induced thermogenesis
ISS
PMID:26166748
Adiponectin Enhances Cold-Induced Browning of Subcutaneous A...
REMOVE
Summary: Sign inversion: this reference reports that adiponectin PROMOTES cold-induced browning, so it cannot support the negative-regulation term.
Reason: GOA cites GO:0120162 (positive) and GO:0120163 (negative) regulation of cold-induced thermogenesis to the SAME two references, PMID:24531262 and PMID:26166748 - a full 2x2 cross-product. The same evidence cannot support a proposition and its negation. Reading the two papers assigns each to exactly one direction: PMID:24531262 (Diabetologia 2014) reports that Adipoq-knockout mice run HOTTER and express MORE UCP1, i.e. adiponectin SUPPRESSES thermogenesis; PMID:26166748 (Cell Metab 2015) reports that the cold-induced thermogenic program is IMPAIRED in adiponectin-knockout mice, i.e. adiponectin PROMOTES it. The same cross-product is present on mouse Adipoq (Q60994) as four IMP rows from the same two references, so the defect originates upstream at MGI and reaches human through these ISS transfers. This particular row pairs the NEGATIVE term with the paper that demonstrates enhancement, and is removable on the reference's own conclusion. The same paper correctly supports the GO:0120162 row, which is kept.
Propagation Review
Root cause: SOURCE BAD
Failure modes: REGULATORY SIGN INVERSION SOURCE MISCITATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq carries the identical cross-product: GO:0120162 and GO:0120163 each by IMP from BOTH PMID:24531262 and PMID:26166748
Supporting Evidence:
PMID:26166748
Chronic cold exposure-induced accumulation of M2 macrophages, activation of beige cells, and thermogenic program were markedly impaired in scWAT of adiponectin knockout (ADN KO) mice
GO:0005576 extracellular region
HDA
PMID:27068509
Extracellular matrix remodelling in response to venous hyper...
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree.
GO:1904706 negative regulation of vascular associated smooth muscle cell proliferation
IDA
PMID:12070119
Adipocyte-derived plasma protein adiponectin acts as a plate...
KEEP AS NON CORE
Summary: Adiponectin inhibits proliferation of vascular smooth muscle cells.
Reason: PMID:12070119 showed physiological concentrations of adiponectin suppress PDGF-BB-stimulated proliferation of human aortic smooth muscle cells. A direct consequence of PDGF-BB sequestration, and one of the better-supported vascular effects.
Supporting Evidence:
PMID:12070119
Physiological concentrations of adiponectin significantly suppressed both proliferation and migration of HASMCs stimulated with platelet-derived growth factor (PDGF)-BB
GO:1904753 negative regulation of vascular associated smooth muscle cell migration
IDA
PMID:12070119
Adipocyte-derived plasma protein adiponectin acts as a plate...
KEEP AS NON CORE
Summary: Adiponectin inhibits migration of vascular smooth muscle cells.
Reason: The migration half of the same PMID:12070119 experiment, measured in a Boyden chamber. Downstream of PDGF-BB binding.
Supporting Evidence:
PMID:12070119
Cell migration assay was performed using a Boyden chamber
GO:0010875 positive regulation of cholesterol efflux
IDA
PMID:18703020
Adiponectin prevents atherosclerosis by increasing cholester...
KEEP AS NON CORE
Summary: Adiponectin raises ABCA1-dependent cholesterol efflux from macrophages.
Reason: PMID:18703020 showed APN upregulates ABCA1, LXRalpha and PPARgamma and increases apoA-I-mediated efflux in human macrophages. A downstream transcriptional consequence of adiponectin signalling.
GO:0009986 cell surface
IDA
PMID:10982546
Adiponectin, an adipocyte-derived plasma protein, inhibits e...
KEEP AS NON CORE
Summary: Adiponectin docks saturably on the surface of endothelial cells, which is where its receptors are; a real but secondary location.
Reason: PMID:10982546 measured biotinylated adiponectin binding to human aortic endothelial cells by cell ELISA and found specific, saturable binding. For a secreted ligand, 'cell surface' records where it accumulates on the responding cell, which is a consequence of the receptor-binding function rather than a separate one.
Supporting Evidence:
PMID:10982546
Adiponectin specifically bound to HAECs in a saturable manner and inhibited TNF-alpha-induced mRNA expression of monocyte adhesion molecules
GO:0005576 extracellular region
TAS
Reactome:R-HSA-1183058
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree.
GO:0005576 extracellular region
TAS
Reactome:R-HSA-8848663
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree.
GO:2000584 negative regulation of platelet-derived growth factor receptor-alpha signaling pathway
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Adiponectin dampens PDGF-alpha-receptor signalling downstream of the receptor.
Reason: Worth stating precisely, because PMID:12070119 reports that adiponectin reduced PDGF-AA-stimulated ERK phosphorylation 'without affecting autophosphorylation of PDGF alpha-receptor', and that adiponectin does not bind PDGF-AA. So the alpha-receptor pathway is inhibited at a post-receptor step, unlike the beta-receptor pathway where ligand sequestration acts upstream. The term is about the pathway, which includes downstream steps, so it stands - but it should not be read as ligand sequestration.
Supporting Evidence:
PMID:12070119
without affecting autophosphorylation of PDGF alpha-receptor
GO:2000590 negative regulation of metanephric mesenchymal cell migration
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Adiponectin inhibits PDGF-driven migration of mesangial cells.
Reason: Transferred from mouse Adipoq, whose annotation rests on PMID:19460854, 'Inhibitory effects of adiponectin on platelet-derived growth factor-induced mesangial cell migration'. The term names metanephric MESENCHYMAL cells, the embryonic progenitors, whereas the assay used cultured mesangial cells; GO has no mesangial cell migration term, so this is the closest available fit rather than an error. Mechanistically it is again PDGF antagonism.
GO:0033691 sialic acid binding
IDA
PMID:19855092
Sialic acid modification of adiponectin is not required for ...
REMOVE
Summary: The cited paper shows that adiponectin IS sialylated; it does not show that adiponectin binds sialic acid. The annotation inverts the direction of the relationship.
Reason: GO:0033691 is defined as 'Binding to a sialic acid'. PMID:19855092 is a post-translational-modification study: it maps sialylated O-linked glycans onto Thr residues of adiponectin's own variable domain and shows that removing them accelerates plasma clearance. Adiponectin is the glycoprotein carrying the sialic acid, not a lectin recognising it - the only receptor discussed is the hepatic asialoglycoprotein receptor, which binds desialylated adiponectin, i.e. the traffic runs the other way. UniProt encodes the same facts as CARBOHYD features at Thr-21 and Thr-22 with this exact reference, and a PTM comment, not as a binding activity. Adiponectin has no lectin domain: its two modules are a collagen-like repeat (42-107) and a C1q jelly-roll (108-244). This is the sole annotation PMID:19855092 carries in all of GOA. Note the row has already propagated: mouse Adipoq (Q60994) holds GO:0033691 only by IEA GO_REF:0000107 and ISO GO_REF:0000119, both orthology transfers from this human row, so correcting it here retracts the error in both species.
Supporting Evidence:
PMID:19855092
sialylation occurs on previously unidentified O-linked glycans on Thr residues of the variable domain in human adiponectin
PMID:19855092
plasma clearance of desialylated adiponectin was accelerated compared with that of control adiponectin
GO:2000534 positive regulation of renal albumin absorption
IDA
PMID:18431508
Adiponectin regulates albuminuria and podocyte function in m...
MARK AS OVER ANNOTATED
Summary: The paper measures podocyte permeability to albumin, not renal albumin absorption; the term names the wrong process.
Reason: GO:2000534 sits under GO:0097017 renal protein absorption, defined as proteins 'taken up from the collecting ducts, glomerulus and proximal and distal loops of the nephron'. PMID:18431508 measures something else entirely: it shows that adiponectin-knockout mice have podocyte foot-process effacement and increased albuminuria, and that adiponectin acting through AMPK reduces podocyte PERMEABILITY to albumin. The full text contains no measurement of tubular reabsorption; the words absorption, reabsorption and tubular do not appear in the extracted text at all. So adiponectin restores the glomerular filtration barrier - less albumin crosses it - rather than increasing uptake downstream. GO has no term for regulation of glomerular permeability to albumin, which is very likely why this one was chosen; that gap is filed under proposed_new_terms. Marked over-annotated rather than removed because the phenotype (less urinary albumin) is real and the term is the closest existing fit. Note also that the evidence code is IDA on the human gene while the experiments are an adiponectin-knockout mouse plus cultured podocytes; MGI records the same finding as IMP on mouse Adipoq, which is the more accurate code.
Supporting Evidence:
PMID:18431508
both adiponectin and AMPK activation reduced podocyte permeability to albumin and podocyte dysfunction
GO:2000467 positive regulation of glycogen (starch) synthase activity
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Adiponectin increases glycogen synthase activity in responding tissue.
Reason: ISS transferred from UniProtKB:Q8K3R4, which resolves to rat Adipoq but is an UNREVIEWED TrEMBL entry rather than a Swiss-Prot ortholog record - weaker support than the mouse-derived rows, and worth noting. The effect itself is consistent with adiponectin's insulin-sensitising action.
GO:0032757 positive regulation of interleukin-8 production
IDA
PMID:19524870
Induction of chemokine expression by adiponectin in vitro is...
KEEP AS NON CORE
Summary: HMW adiponectin induces IL-8 in PBMCs and microvascular endothelial cells.
Reason: PMID:19524870 is a careful isoform-resolved study: HMW but not LMW adiponectin induces chemokines. It is retained because it is one of the clearest demonstrations that adiponectin's inflammatory sign is multimer-dependent, and it sits alongside the anti-inflammatory rows without contradicting them.
Supporting Evidence:
PMID:19524870
induced a dose-dependent increase in production of monocyte chemoattractant protein-1 and interleukin-8 by PBMCs and MVECs
GO:0071639 positive regulation of monocyte chemotactic protein-1 production
IDA
PMID:19524870
Induction of chemokine expression by adiponectin in vitro is...
KEEP AS NON CORE
Summary: HMW adiponectin induces MCP-1 in PBMCs and microvascular endothelial cells.
Reason: The MCP-1 half of PMID:19524870's isoform-dependent chemokine result, alongside the IL-8 row. Kept for the same reason: it documents the multimer dependence of adiponectin's inflammatory sign.
Supporting Evidence:
PMID:19524870
LMW adiponectin at the same concentrations did not induce chemokine production in any of the cell types tested
GO:0005515 protein binding
IPI
PMID:12070119
Adipocyte-derived plasma protein adiponectin acts as a plate...
MODIFY
Summary: This is not a screen hit: adiponectin was shown to bind PDGF-BB directly, and growth factor binding says so.
Reason: PMID:12070119 is a hypothesis-led study in human aortic smooth muscle cells. Adiponectin bound 125I-PDGF-BB and blocked its association with the cells, while having no effect on the binding of 125I-PDGF-AA or 125I-HB-EGF - i.e. the interaction was measured and shown to be selective. GO:0019838 growth factor binding ('binding to a growth factor, proteins or polypeptides that stimulate a cell or organism to grow or proliferate') states the informative content that bare protein binding discards, and it is the molecular event underlying this gene's GO:0010642, GO:0070373, GO:1904706 and GO:1904753 rows. UniProt's SUBUNIT records the same for the mouse protein and extends it to HBEGF and FGF2 with multimer-specific affinities, which remains By similarity for human.
Proposed replacements: growth factor binding
Supporting Evidence:
PMID:12070119
Adiponectin specifically bound to (125)I-PDGF-BB and significantly inhibited the association of (125)I-PDGF-BB with HASMCs
GO:0005576 extracellular region
IDA
PMID:15585515
Adiponectin in chronic kidney disease is related more to met...
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree.
GO:0010642 negative regulation of platelet-derived growth factor receptor signaling pathway
IDA
PMID:12070119
Adipocyte-derived plasma protein adiponectin acts as a plate...
ACCEPT
Summary: Adiponectin binds PDGF-BB and blocks its access to the receptor; a directly demonstrated, mechanistically explicit process.
Reason: This is one of the few adiponectin activities with a directly measured molecular mechanism in human cells. PMID:12070119 showed that human adiponectin binds PDGF-BB itself and thereby blocks the ligand from reaching its receptor on human aortic smooth muscle cells, suppressing PDGF beta-receptor autophosphorylation. Ligand sequestration is a direct participation in the negative regulation of that pathway, not a downstream physiological consequence.
Supporting Evidence:
PMID:12070119
Adiponectin specifically bound to (125)I-PDGF-BB and significantly inhibited the association of (125)I-PDGF-BB with HASMCs
GO:0070373 negative regulation of ERK1 and ERK2 cascade
IDA
PMID:12070119
Adipocyte-derived plasma protein adiponectin acts as a plate...
KEEP AS NON CORE
Summary: Adiponectin suppresses growth-factor-induced ERK1/2 signalling in vascular smooth muscle.
Reason: PMID:12070119 showed adiponectin 'strongly and dose-dependently suppressed PDGF-BB-induced p42/44 extracellular signal-related kinase (ERK) phosphorylation'. Mechanistically downstream of the PDGF-BB sequestration recorded by GO:0010642.
Supporting Evidence:
PMID:12070119
Adiponectin strongly and dose-dependently suppressed PDGF-BB-induced p42/44 extracellular signal-related kinase (ERK) phosphorylation and PDGF beta-receptor autophosphorylation
GO:0005576 extracellular region
IDA
PMID:17327472
Effects of adiponectin on the renal sympathetic nerve activi...
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree.
GO:0010906 regulation of glucose metabolic process
IDA
PMID:17327472
Effects of adiponectin on the renal sympathetic nerve activi...
ACCEPT
Summary: Regulation of glucose metabolism is a genuine core action of adiponectin, though this particular reference is a poor citation for it.
Reason: The term is right: suppression of hepatic glucose output is adiponectin's best-characterised metabolic effect (PMID:11748271, PMID:12368907), and this is the term the role-conflated GO:0006006 rows should be modified to. The citation is the weak part - PMID:17327472 measured renal sympathetic nerve activity and blood pressure in rats, and mentions glucose metabolism only as background in its first sentence. The full text is not available here, so per GO curation practice the curator's reading is not overruled; the concern is recorded in this reference's reference_review instead. Accepted on the strength of the term, with the citation flagged.
GO:0045776 negative regulation of blood pressure
IDA
PMID:17327472
Effects of adiponectin on the renal sympathetic nerve activi...
KEEP AS NON CORE
Summary: Adiponectin lowers blood pressure, at least partly via reduced sympathetic outflow.
Reason: PMID:17327472 showed dose-dependent suppression of blood pressure by both intravenous and intracerebroventricular adiponectin in rats, abolished by suprachiasmatic-nucleus lesions. A whole-organism physiological outcome, several steps from the molecular function.
GO:0050728 negative regulation of inflammatory response
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Adiponectin is broadly anti-inflammatory.
Reason: A general statement supported by the endothelial adhesion molecule, TNF and macrophage results. Kept as the umbrella term for a set of downstream immune effects, none of which is the gene's molecular function.
GO:0050805 negative regulation of synaptic transmission
IDA
PMID:17327472
Effects of adiponectin on the renal sympathetic nerve activi...
MARK AS OVER ANNOTATED
Summary: The cited experiment records renal sympathetic nerve firing and blood pressure, not transmission across a synapse.
Reason: GO:0050805 is defined as reducing 'synaptic transmission, the process of communication from a neuron to a target (neuron, muscle, or secretory cell) across a synapse'. PMID:17327472 injected adiponectin intravenously and intracerebroventricularly into anaesthetised rats and measured renal sympathetic nerve activity and blood pressure, with the effect abolished by suprachiasmatic-nucleus lesions. That is efferent autonomic outflow, one level of organisation above synaptic transmission, and no synaptic measurement was made. GO has no term for regulation of sympathetic nervous system activity - a search of the ontology returns only developmental terms (GO:0048485, GO:0061549, GO:0097490-2, GO:1903045) - so there is nothing to MODIFY to; the gap is filed under proposed_new_terms. The blood-pressure half of the same experiment is already captured by the GO:0045776 row.
Supporting Evidence:
PMID:17327472
Both iv and LCV injections of adiponectin induced dose-dependent suppressions of RSNA and b/p
GO:0071466 cellular response to xenobiotic stimulus
ISS
GO_REF:0000024
MARK AS OVER ANNOTATED
Summary: The underlying observation is that a drug INDUCES adiponectin, which makes adiponectin the output of the response rather than a participant in it.
Reason: The mouse annotation this ISS transfers from is GO:0071466 IDA on Q60994 from PMID:19109165, 'The peroxisome proliferator-activated receptor gamma agonist rosiglitazone ameliorates murine lupus by induction of adiponectin'. The datum is that rosiglitazone raises adiponectin expression. Annotating ADIPOQ to 'cellular response to xenobiotic stimulus' therefore inverts the roles: adiponectin is what the responding adipocyte produces, not the machinery by which it detects or handles the xenobiotic. Marked over-annotated rather than removed because the transcriptional response is real and a curator reading only the expression change would reasonably reach for this term.
Propagation Review
Root cause: SOURCE BAD
Failure modes: ROLE CONFLATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq GO:0071466 IDA from PMID:19109165, a rosiglitazone-induction study
GO:0005102 signaling receptor binding
ISS
GO_REF:0000024
ACCEPT
Summary: Adiponectin binds its signalling receptors; kept as the general-grain statement of the core ligand function.
Reason: Adiponectin binds AdipoR1 and AdipoR2 (PMID:12802337) and T-cadherin (PMID:15210937), so the term is correct and experimentally well founded. It is worth being precise about how it relates to the gene's other molecular-function rows, because the two look redundant and are not: GO:0005102 signaling receptor binding is a BINDING term, whereas GO:0005179 hormone activity sits in the ACTIVITY branch, under GO:0048018 receptor ligand activity -> GO:0140677 molecular function activator activity -> GO:0098772 molecular function regulator activity. GO:0005102 is NOT among GO:0005179's is_a/part_of ancestors (verified against QuickGO and recorded under term_relations in ADIPOQ-bioinformatics/results.json). The two rows therefore state different things - that adiponectin physically engages a receptor, and that engaging it activates the receptor - and neither subsumes the other.
GO:0005783 endoplasmic reticulum
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Adiponectin transits and is assembled in the ER, but the ER is a biosynthetic waypoint rather than the site of its function.
Reason: Adiponectin's multimer assembly, disulfide bonding and lysine hydroxylation/glycosylation occur in the endoplasmic reticulum before secretion, so the localisation is real and mechanistically important - UniProt PTM notes that hydroxylation and glycosylation of the collagen-domain lysines are 'critically involved in regulating the formation and/or secretion of HMW complexes'. It is non-core because the protein acts after secretion, on other cells.
GO:0006006 glucose metabolic process
ISS
GO_REF:0000024
MODIFY
Summary: Adiponectin regulates this process in responding cells; it does not carry it out.
Reason: GO:0006006 glucose metabolic process is a direct-participation term - 'the chemical reactions and pathways involving glucose'. Adiponectin is a secreted hormone: it does not itself carry out glucose or fatty-acid metabolism, it changes the rate at which responding cells do so, via AdipoR1/AdipoR2 and AMPK. PMID:12368907 is explicit that the effect is mediated by AMPK in the responding myocyte or hepatocyte, and that blocking AMPK abolishes it. GO:0010906 regulation of glucose metabolic process states what adiponectin does, and ADIPOQ already carries it, so this is a correction of grain rather than a loss of content.
Propagation Review
Root cause: TERM SCOPING PROBLEM
Failure modes: ROLE CONFLATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq; holds GO:0006006 itself by IDA/IMP, so the same conflation exists upstream at MGI
GO:0006635 fatty acid beta-oxidation
ISS
GO_REF:0000024
MODIFY
Summary: Adiponectin regulates this process in responding cells; it does not carry it out.
Reason: Adiponectin is not an enzyme of beta-oxidation and has no catalytic activity of any kind; it raises the rate of fatty-acid oxidation in muscle and liver by activating AMPK downstream of AdipoR1/AdipoR2. Adiponectin is a secreted hormone: it does not itself carry out glucose or fatty-acid metabolism, it changes the rate at which responding cells do so, via AdipoR1/AdipoR2 and AMPK. GO:0046321 positive regulation of fatty acid oxidation states the measured effect at the correct grain and preserves the direction, which the bare metabolic term does not.
Propagation Review
Root cause: TERM SCOPING PROBLEM
Failure modes: ROLE CONFLATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq; holds GO:0006635 (IMP) and GO:0019395 (IDA) itself
GO:0019395 fatty acid oxidation
ISS
GO_REF:0000024
MODIFY
Summary: Adiponectin regulates this process in responding cells; it does not carry it out.
Reason: Adiponectin is not an enzyme of beta-oxidation and has no catalytic activity of any kind; it raises the rate of fatty-acid oxidation in muscle and liver by activating AMPK downstream of AdipoR1/AdipoR2. Adiponectin is a secreted hormone: it does not itself carry out glucose or fatty-acid metabolism, it changes the rate at which responding cells do so, via AdipoR1/AdipoR2 and AMPK. GO:0046321 positive regulation of fatty acid oxidation states the measured effect at the correct grain and preserves the direction, which the bare metabolic term does not.
Propagation Review
Root cause: TERM SCOPING PROBLEM
Failure modes: ROLE CONFLATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq; holds GO:0006635 (IMP) and GO:0019395 (IDA) itself
GO:0030336 negative regulation of cell migration
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Adiponectin restrains migration of vascular smooth muscle and mesangial cells.
Reason: Transferred by ISS from mouse Adipoq, whose GO:0030336 IDA comes from PMID:19460854 on PDGF-induced mesangial cell migration. Mechanistically this is the same PDGF-BB sequestration captured more precisely by GO:0010642 and GO:1904753; kept as the general vascular-remodelling statement.
GO:0032869 cellular response to insulin stimulus
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Adipocytes and adiponectin-responsive tissues respond to insulin.
Reason: ISS transfer from mouse. Adiponectin sensitises tissues to insulin and its own expression is insulin-responsive; the term is broad and peripheral to the gene's own activity.
GO:0042593 glucose homeostasis
ISS
GO_REF:0000024
ACCEPT
Summary: Maintenance of blood glucose is adiponectin's principal systemic role.
Reason: Adiponectin lowers glucose by suppressing hepatic glucose output and increasing peripheral glucose disposal; knockout mice are glucose intolerant and adiponectin replacement reverses insulin resistance. This is a core organismal process for the gene.
GO:0043123 positive regulation of canonical NF-kappaB signal transduction
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Hexameric and HMW adiponectin activate NF-kB, in contrast to the trimer.
Reason: This is NOT a contradiction of the GO:0043124 negative row. PMID:12087086 and PMID:14522956 show that hexamer and HMW Acrp30 activate NF-kB in C2C12 myocytes while trimers and the globular fragment do not, whereas PMID:10982546 shows adiponectin suppressing TNF-INDUCED NF-kB in endothelium. The sign depends on multimer state and cell type, and both rows are correct.
Supporting Evidence:
file:human/ADIPOQ/ADIPOQ-deep-research-affinage.md
Hexameric and higher molecular weight (HMW) isoforms of Acrp30 activate NF-ΞΊB in C2C12 myocytes via phosphorylation and degradation of IΞΊB-alpha, whereas trimeric Acrp30 and globular domain (gAcrp30) do not
GO:0043407 negative regulation of MAP kinase activity
ISS
GO_REF:0000024
KEEP AS NON CORE
Summary: Adiponectin dampens MAP kinase activation by growth factors.
Reason: ISS transfer from mouse; mechanistically the same result as the human GO:0070373 IDA row, where adiponectin suppresses PDGF-BB-induced ERK phosphorylation. Kept at the general grain.
GO:0050873 brown fat cell differentiation
ISS
GO_REF:0000024
MODIFY
Summary: Adiponectin promotes browning of adipose tissue; it does not itself undergo brown fat cell differentiation.
Reason: GO:0050873 is the differentiation process itself - 'the process in which a relatively unspecialized cell acquires specialized features of a brown adipocyte'. Adiponectin is a secreted product of white adipocytes that acts on other cells; the reported effect is that cold-induced adiponectin recruits M2 macrophages via T-cadherin and thereby activates beige cells (PMID:26166748). GO:0090336 positive regulation of brown fat cell differentiation states that correctly and adds the direction. Note this is the same claim whose citation is disputed by PMID:24531262; the regulation term is used here because the ISS transfer is anchored on the browning result.
Propagation Review
Root cause: TERM SCOPING PROBLEM
Failure modes: ROLE CONFLATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq; GO:0050873 IDA from PMID:18492766, a paper about a different adipokine in which adiponectin serves as a differentiation marker
GO:0010745 negative regulation of macrophage derived foam cell differentiation
IDA
PMID:11222466
Adipocyte-derived plasma protein, adiponectin, suppresses li...
KEEP AS NON CORE
Summary: Adiponectin suppresses the conversion of macrophages into lipid-laden foam cells.
Reason: PMID:11222466 showed that adiponectin lowers class A scavenger receptor expression and lipid uptake in human monocyte-derived macrophages. Real and human, but this is a downstream cellular consequence of receptor signalling, not a molecular action of the protein.
GO:0010804 negative regulation of tumor necrosis factor-mediated signaling pathway
IDA
PMID:10604883
Novel modulator for endothelial adhesion molecules: adipocyt...
KEEP AS NON CORE
Summary: Adiponectin blunts TNF signalling in endothelium.
Reason: Downstream physiology: adiponectin pretreatment suppresses TNF-induced IkB-alpha phosphorylation and adhesion molecule expression. Anti-TNF action is a well-established adiponectin effect but is mediated through its own receptor pathway.
Supporting Evidence:
PMID:10604883
Physiological concentrations of adiponectin dose-dependently inhibited TNF-alpha-induced THP-1 adhesion and expression of VCAM-1, E-selectin, and ICAM-1 on HAECs
GO:0005102 signaling receptor binding
ISS
PMID:12368907
Adiponectin stimulates glucose utilization and fatty-acid ox...
ACCEPT
Summary: Adiponectin binds its signalling receptors; kept as the general-grain statement of the core ligand function.
Reason: Adiponectin binds AdipoR1 and AdipoR2 (PMID:12802337) and T-cadherin (PMID:15210937), so the term is correct and experimentally well founded. It is worth being precise about how it relates to the gene's other molecular-function rows, because the two look redundant and are not: GO:0005102 signaling receptor binding is a BINDING term, whereas GO:0005179 hormone activity sits in the ACTIVITY branch, under GO:0048018 receptor ligand activity -> GO:0140677 molecular function activator activity -> GO:0098772 molecular function regulator activity. GO:0005102 is NOT among GO:0005179's is_a/part_of ancestors (verified against QuickGO and recorded under term_relations in ADIPOQ-bioinformatics/results.json). The two rows therefore state different things - that adiponectin physically engages a receptor, and that engaging it activates the receptor - and neither subsumes the other.
GO:0005125 cytokine activity
NAS
PMID:12611609
The role of the novel adipocyte-derived hormone adiponectin ...
KEEP AS NON CORE
Summary: Adiponectin has genuine cytokine-like actions on myeloid cells, but its defining molecular function is endocrine, not cytokine.
Reason: GO:0005125 is defined as 'the activity of a soluble extracellular gene product that interacts with a receptor to effect a change in the activity of the receptor to control the survival, growth, differentiation and effector function of tissues and cells'. Adiponectin does act on macrophages and myelomonocytic progenitors (PMID:10961870), so the term is not wrong. It is retained as non-core because the evidence is NAS - an author statement in a review article, the only two annotations that reference carries in all of GOA - and because the gene's characterised identity is a circulating metabolic hormone rather than an immune mediator.
GO:0006006 glucose metabolic process
ISS
PMID:12368907
Adiponectin stimulates glucose utilization and fatty-acid ox...
MODIFY
Summary: Adiponectin regulates this process in responding cells; it does not carry it out.
Reason: GO:0006006 glucose metabolic process is a direct-participation term - 'the chemical reactions and pathways involving glucose'. Adiponectin is a secreted hormone: it does not itself carry out glucose or fatty-acid metabolism, it changes the rate at which responding cells do so, via AdipoR1/AdipoR2 and AMPK. PMID:12368907 is explicit that the effect is mediated by AMPK in the responding myocyte or hepatocyte, and that blocking AMPK abolishes it. GO:0010906 regulation of glucose metabolic process states what adiponectin does, and ADIPOQ already carries it, so this is a correction of grain rather than a loss of content.
Propagation Review
Root cause: TERM SCOPING PROBLEM
Failure modes: ROLE CONFLATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq; holds GO:0006006 itself by IDA/IMP, so the same conflation exists upstream at MGI
GO:0006635 fatty acid beta-oxidation
ISS
PMID:12368907
Adiponectin stimulates glucose utilization and fatty-acid ox...
MODIFY
Summary: Adiponectin regulates this process in responding cells; it does not carry it out.
Reason: Adiponectin is not an enzyme of beta-oxidation and has no catalytic activity of any kind; it raises the rate of fatty-acid oxidation in muscle and liver by activating AMPK downstream of AdipoR1/AdipoR2. Adiponectin is a secreted hormone: it does not itself carry out glucose or fatty-acid metabolism, it changes the rate at which responding cells do so, via AdipoR1/AdipoR2 and AMPK. GO:0046321 positive regulation of fatty acid oxidation states the measured effect at the correct grain and preserves the direction, which the bare metabolic term does not.
Propagation Review
Root cause: TERM SCOPING PROBLEM
Failure modes: ROLE CONFLATION
Sources checked:
UniProtKB:Q60994
mouse Adipoq; holds GO:0006635 (IMP) and GO:0019395 (IDA) itself
Supporting Evidence:
PMID:12368907
stimulation of glucose utilization and fatty-acid oxidation by Ad occurs through activation of AMPK
GO:0009749 response to glucose
ISS
PMID:12368907
Adiponectin stimulates glucose utilization and fatty-acid ox...
KEEP AS NON CORE
Summary: Adipocyte adiponectin expression and secretion track glucose availability.
Reason: A response-to-stimulus term transferred by ISS from mouse. It describes the regulation of adiponectin rather than an action of adiponectin, so it is retained only as peripheral context.
GO:0009967 positive regulation of signal transduction
ISS
PMID:12368907
Adiponectin stimulates glucose utilization and fatty-acid ox...
MODIFY
Summary: Too general to be informative for a gene whose own signalling pathway has a dedicated GO term.
Reason: GO:0009967 positive regulation of signal transduction says only that adiponectin turns some pathway up. The pathway in question has its own term, GO:0033211 adiponectin-activated signaling pathway, which ADIPOQ already carries by IDA and ISS. The generic term is not the least common ancestor of a heterogeneous donor set here - there is a single, named pathway - so replacing it loses nothing and gains specificity.
GO:0030853 negative regulation of granulocyte differentiation
IDA
PMID:10961870
Adiponectin, a new member of the family of soluble defense c...
KEEP AS NON CORE
Summary: Adiponectin inhibits granulocyte differentiation from myelomonocytic progenitors.
Reason: PMID:10961870 showed adiponectin suppresses growth of myelomonocytic progenitors and mature macrophage functions. A haematopoietic effect well outside the gene's metabolic core.
GO:0032720 negative regulation of tumor necrosis factor production
IDA
PMID:10961870
Adiponectin, a new member of the family of soluble defense c...
KEEP AS NON CORE
Summary: Adiponectin lowers macrophage TNF production.
Reason: PMID:10961870 found adiponectin inhibits lipopolysaccharide-induced TNF production by cultured macrophages. UniProt's FUNCTION records the same: 'Antagonizes TNF by negatively regulating its expression in various tissues such as liver and macrophages'. Downstream of receptor signalling.
GO:0033034 positive regulation of myeloid cell apoptotic process
IDA
PMID:10961870
Adiponectin, a new member of the family of soluble defense c...
KEEP AS NON CORE
Summary: Adiponectin promotes apoptosis of myelomonocytic cells.
Reason: PMID:10961870 observed subdiploid peaks and oligonucleosomal DNA fragmentation in acute myelomonocytic leukaemia lines treated with adiponectin. A cell-type-restricted downstream effect.
GO:0034115 negative regulation of heterotypic cell-cell adhesion
IDA
PMID:10604883
Novel modulator for endothelial adhesion molecules: adipocyt...
KEEP AS NON CORE
Summary: Adiponectin reduces monocyte adhesion to endothelium.
Reason: PMID:10604883 showed adiponectin dose-dependently inhibits TNF-induced THP-1 adhesion to human aortic endothelial cells. Downstream of the NF-kB suppression captured by GO:0043124.
GO:0034383 low-density lipoprotein particle clearance
IDA
PMID:11222466
Adipocyte-derived plasma protein, adiponectin, suppresses li...
KEEP AS NON CORE
Summary: Adiponectin influences clearance of LDL particles by macrophages.
Reason: Derived from the same scavenger-receptor experiments in PMID:11222466. A downstream lipid-handling consequence.
GO:0034612 response to tumor necrosis factor
IDA
PMID:10604883
Novel modulator for endothelial adhesion molecules: adipocyt...
KEEP AS NON CORE
Summary: Adiponectin-treated endothelial cells respond differently to TNF.
Reason: A response-to-stimulus term from PMID:10604883. It describes the experimental setting - TNF challenge - rather than an adiponectin activity, so it is kept only as context.
GO:0042593 glucose homeostasis
ISS
PMID:12368907
Adiponectin stimulates glucose utilization and fatty-acid ox...
ACCEPT
Summary: Maintenance of blood glucose is adiponectin's principal systemic role.
Reason: Adiponectin lowers glucose by suppressing hepatic glucose output and increasing peripheral glucose disposal; knockout mice are glucose intolerant and adiponectin replacement reverses insulin resistance. This is a core organismal process for the gene.
GO:0042803 protein homodimerization activity
IPI
PMID:12021245
Paracrine regulation of fat cell formation in bone marrow cu...
MODIFY
Summary: Adiponectin's fundamental self-associated unit is a TRIMER, not a dimer; the parent term states the fact without the wrong stoichiometry.
Reason: GO:0042803 is defined as 'Binding to an identical protein to form a homodimer'. No homodimeric adiponectin species has been described. UniProt SUBUNIT records trimers, hexamers and 12- to 18-mers, and the original characterisation of the human protein describes a collagen-like domain 'through which they form homo-trimers, which further combine to make oligomeric complexes'. Generalising to GO:0042802 identical protein binding, the direct parent, keeps everything the evidence supports and drops the dimer claim. The trimer itself is proposed separately as GO:0070207 protein homotrimerization, which is the term that actually states the stoichiometry.
Proposed replacements: identical protein binding
Supporting Evidence:
PMID:8947845
possessing a collagen-like domain through which they form homo-trimers, which further combine to make oligomeric complexes
GO:0043124 negative regulation of canonical NF-kappaB signal transduction
IDA
PMID:10982546
Adiponectin, an adipocyte-derived plasma protein, inhibits e...
KEEP AS NON CORE
Summary: Adiponectin suppresses TNF-induced NF-kB activation in endothelial cells through cAMP/PKA.
Reason: PMID:10982546 showed adiponectin blocks TNF-induced IkB-alpha phosphorylation without affecting JNK, p38 or Akt, and that the effect is cAMP-dependent. UniProt FUNCTION records it as 'Inhibits endothelial NF-kappa-B signaling through a cAMP-dependent pathway'. Well founded but downstream of receptor engagement.
Supporting Evidence:
PMID:10982546
Adiponectin suppressed TNF-alpha-induced IkappaB-alpha phosphorylation and subsequent NF-kappaB activation
GO:0045599 negative regulation of fat cell differentiation
IDA
PMID:12021245
Paracrine regulation of fat cell formation in bone marrow cu...
ACCEPT
Summary: Adiponectin restrains adipocyte differentiation in a paracrine negative-feedback loop.
Reason: Recombinant adiponectin blocks adipogenesis in bone-marrow stromal cultures through a COX-2/prostaglandin-dependent paracrine loop, establishing preadipocytes as direct adiponectin targets. Combined with the gene's identity as the adipocyte's own secreted product, this is a well-founded feedback function rather than an incidental phenotype.
Supporting Evidence:
PMID:12021245
Recombinant adiponectin blocked fat cell formation in long-term bone marrow cultures and inhibited the differentiation of cloned stromal preadipocytes
GO:0045650 negative regulation of macrophage differentiation
IDA
PMID:10961870
Adiponectin, a new member of the family of soluble defense c...
KEEP AS NON CORE
Summary: Adiponectin inhibits macrophage differentiation.
Reason: From PMID:10961870's myelomonocytic experiments. A haematopoietic effect, downstream and outside the core.
GO:0045721 negative regulation of gluconeogenesis
ISS
PMID:12368907
Adiponectin stimulates glucose utilization and fatty-acid ox...
ACCEPT
Summary: Suppression of hepatic gluconeogenesis is adiponectin's best-characterised metabolic action.
Reason: PMID:11748271 showed Acrp30 infusion reduces endogenous glucose production by about 65% and halves hepatic PEPCK and G6Pase mRNA without affecting peripheral glucose uptake, glycolysis or glycogen synthesis. Regulation, not participation, and therefore the correct grain.
Supporting Evidence:
PMID:12368907
reduction of molecules involved in gluconeogenesis in the liver, and reduction of glucose levels in vivo
GO:0045923 positive regulation of fatty acid metabolic process
ISS
PMID:12368907
Adiponectin stimulates glucose utilization and fatty-acid ox...
ACCEPT
Summary: Adiponectin increases fatty-acid metabolism in muscle and liver.
Reason: The regulation-grade counterpart of the role-conflated GO:0006635 and GO:0019395 rows, and the term those should be read alongside. Supported by AMPK-dependent stimulation of fatty-acid oxidation.
GO:0046326 positive regulation of D-glucose import across plasma membrane
ISS
PMID:12368907
Adiponectin stimulates glucose utilization and fatty-acid ox...
ACCEPT
Summary: Adiponectin stimulates glucose uptake in muscle.
Reason: PMID:12368907 showed adiponectin stimulates glucose uptake and lactate production in myocytes through AMPK, and PMID:16622416 traced GLUT4 membrane translocation to APPL1-Rab5. A core metabolic action, correctly expressed as regulation.
Supporting Evidence:
PMID:12368907
Ad stimulates phosphorylation of acetyl coenzyme A carboxylase (ACC), fatty-acid oxidation, glucose uptake and lactate production in myocytes
GO:0050728 negative regulation of inflammatory response
NAS
PMID:12611609
The role of the novel adipocyte-derived hormone adiponectin ...
KEEP AS NON CORE
Summary: Adiponectin is broadly anti-inflammatory.
Reason: A general statement supported by the endothelial adhesion molecule, TNF and macrophage results. Kept as the umbrella term for a set of downstream immune effects, none of which is the gene's molecular function.
GO:0050765 negative regulation of phagocytosis
IDA
PMID:10961870
Adiponectin, a new member of the family of soluble defense c...
KEEP AS NON CORE
Summary: Adiponectin suppresses macrophage phagocytosis via the C1q receptor C1qRp.
Reason: PMID:10961870 showed adiponectin inhibits phagocytic activity and that this was abrogated by anti-C1qRp antibody, implicating CD93 as a receptor. Notable as one of the few receptor-resolved adiponectin effects, but a downstream immune process.
Supporting Evidence:
PMID:10961870
Suppression of phagocytosis by adiponectin is mediated by one of the complement C1q receptors, C1qRp
GO:0005576 extracellular region
IDA
PMID:10403784
The human apM-1, an adipocyte-specific gene linked to the fa...
ACCEPT
Summary: Adiponectin is a secreted plasma protein; the extracellular region is its core location.
Reason: UniProt SUBCELLULAR LOCATION is 'Secreted' with experimental evidence, and TISSUE SPECIFICITY records that it is 'Synthesized exclusively by adipocytes and secreted into plasma'. Circulating concentrations are 5-30 ug/ml. Multiple independent experimental rows agree.
GO:0006091 generation of precursor metabolites and energy
TAS
PMID:10095105
Organization of the gene for gelatin-binding protein (GBP28)...
REMOVE
Summary: The cited paper is a gene-structure report that performs no metabolic experiment, and the term is a direct-participation term for a protein with no catalytic activity.
Reason: PMID:10095105 is 'Organization of the gene for gelatin-binding protein (GBP28)': it maps ADIPOQ to 3q27 by FISH and describes a 16 kb, three-exon gene lacking a TATA box. It contains no assay of energy metabolism. The projection test shows this reference carries exactly one annotation in all of GOA - this one - so it is not a bulk import that swept ADIPOQ up; it is a single unsupported TAS, assigned by PINC. Separately, GO:0006091 'generation of precursor metabolites and energy' describes the reactions themselves, and adiponectin is a hormone that modulates them in other cells rather than performing them. Mouse Adipoq carries no counterpart row, so nothing depends on this one.
Propagation Review
Root cause: SOURCE BAD
Failure modes: SOURCE MISCITATION ROLE CONFLATION
Sources checked:
PMID:10095105
gene-structure paper; its only GOA annotation is this row
GO:0070207 protein homotrimerization
IDA
PMID:8947845
Isolation and characterization of GBP28, a novel gelatin-bin...
NEW
Summary: Adiponectin's obligate assembly unit is a homotrimer, and GO has a term that states exactly this; nothing in the current record captures the stoichiometry.
Reason: The gene's whole signalling logic runs through assembly state - trimers activate AMPK in skeletal muscle, hexamers and HMW complexes activate NF-kB and are the only species T-cadherin binds - yet GOA records self-association only as GO:0042802 identical protein binding (IEA) and GO:0042803 protein homodimerization activity, the latter naming a stoichiometry adiponectin does not adopt. GO:0070207 is defined as 'the formation of a protein homotrimer, a macromolecular structure consisting of three noncovalently associated identical subunits', which matches UniProt's SUBUNIT description of a non-covalent collagen-domain triple helix precisely. PMID:8947845 characterised the human plasma protein and described this architecture directly.
Supporting Evidence:
PMID:8947845
possessing a collagen-like domain through which they form homo-trimers, which further combine to make oligomeric complexes
PMID:14522956
substitution of Cys22 with alanine led exclusively to trimers

Core Functions

Adipocyte-derived endocrine hormone. Adiponectin is secreted into plasma and acts on distant tissues through the seven-transmembrane receptors AdipoR1 and AdipoR2 and the GPI-anchored receptor T-cadherin, initiating the adiponectin-activated signalling pathway and, through AMPK and PPAR-alpha, raising fatty-acid oxidation and glucose uptake in muscle while suppressing hepatic gluconeogenesis.

Supporting Evidence:
  • PMID:12368907
    stimulation of glucose utilization and fatty-acid oxidation by Ad occurs through activation of AMPK
  • PMID:12802337
    they serve as receptors for globular and full-length adiponectin

Growth-factor sequestration. Adiponectin binds PDGF-BB directly and selectively - it does not bind PDGF-AA - preventing the ligand from engaging its receptor on vascular smooth muscle cells and thereby suppressing PDGF beta-receptor autophosphorylation, ERK1/2 activation, proliferation and migration. This is the one adiponectin activity with a directly measured, ligand-level molecular mechanism in human cells.

Supporting Evidence:
  • PMID:12070119
    Adiponectin specifically bound to (125)I-PDGF-BB and significantly inhibited the association of (125)I-PDGF-BB with HASMCs

Homomultimer assembly. Three adiponectin chains associate through a triple helix of their collagen-like domains and hydrophobic contacts in the globular C1q heads to form the obligate trimer; trimers then join through interchain disulfide bonds into hexamers and 12- to 18-mer high-molecular-weight complexes. The assembly state is functionally decisive - trimers activate AMPK in muscle, hexamers and HMW species activate NF-kB and are the only forms that bind T-cadherin - so self-association is part of the gene's function rather than incidental. GO:0042802 is a deliberately unspecific term, and it is used here because it is the most specific CORRECT molecular function available: GO:0042803 protein homodimerization activity names a stoichiometry adiponectin does not adopt, and GO:0070207 protein homotrimerization - the term that does state it - exists only in the biological process branch. A search of the ontology returns no molecular function term for homotrimerisation or homooligomerisation, so the imprecision is the ontology's, not the annotation's.

Molecular Function:
identical protein binding
Directly Involved In:
Cellular Locations:
Supporting Evidence:
  • PMID:8947845
    possessing a collagen-like domain through which they form homo-trimers, which further combine to make oligomeric complexes
  • PMID:14522956
    Thus, trimeric and HMW/hexameric Acrp30 activate different signal transduction pathways

References

Loading supporting content…

Download this section (compressed HTML)

Suggested Questions for Experts

Q: GOA and MGI both attach GO:0120162 (positive) and GO:0120163 (negative) regulation of cold-induced thermogenesis to BOTH PMID:24531262 and PMID:26166748 - a full 2x2 cross-product in which each paper supports only one direction. Mouse Adipoq carries four IMP rows over the same two references and human ADIPOQ four ISS/IEA rows derived from them. Can MGI reduce this to one correctly-cited row per direction, after which the human ISS and Compara rows will follow?

Q: GO records no interaction between adiponectin and T-cadherin (CDH13, P55290) on either gene: CDH13 has 60 annotations and none names Q15848 in WITH/FROM, and ADIPOQ's 61 GO:0005515 partners do not include it. T-cadherin is the established receptor for hexameric and HMW adiponectin (PMID:15210937) and mediates the M2-macrophage recruitment in PMID:26166748. AdipoR2 (Q86V24) is likewise absent from both records while AdipoR1 has a reciprocal pair. Should the two missing receptors be curated?

Q: GO:0045296 cadherin binding is defined as 'binding to cadherin, a type I membrane protein involved in cell adhesion'. T-cadherin is GPI-anchored, not a type I membrane protein, so the definition as written excludes the one cadherin that is a hormone receptor. Should the definition be broadened to the cadherin superfamily by domain content rather than by membrane topology?

Q: GO:0033691 sialic acid binding on ADIPOQ (IDA, PMID:19855092) records that adiponectin IS sialylated, not that it binds sialic acid; the same reference is used correctly by UniProt as CARBOHYD features. Mouse Adipoq has acquired the term by IEA and ISO from this human row, so it is the sole origin. Can it be retracted in both species?

Q: Adiponectin's activity is multimer-dependent in a way GO cannot currently express: trimers activate AMPK but not NF-kB, hexamers and HMW complexes do the reverse and alone bind T-cadherin, and HMW but not LMW induces MCP-1 and IL-8. Several annotations on this gene are isoform-specific in that sense while carrying no marker of it. Is there an annotation-extension convention for oligomeric state, analogous to the isoform field for splice variants?

Q: UniProt's SUBUNIT states that LMW, MMW and HMW complexes bind HBEGF and that HMW binds FGF2, with distinct affinities, but tags the whole statement ECO:0000250|UniProtKB:Q60994 (By similarity from mouse). The one human measurement, PMID:12070119, found no effect of adiponectin on HB-EGF binding to smooth muscle cells. Is there direct human evidence for HBEGF and FGF2 binding, or should the human entry be qualified?

Suggested Experiments

Experiment: Resolve the cold-thermogenesis contradiction by comparing the two knockout lines head to head. PMID:24531262 and PMID:26166748 used different Adipoq-null alleles, different cold protocols (acute 4 C challenge versus chronic intermittent exposure) and different depots (interscapular BAT versus subcutaneous WAT beiging). Run both protocols on both lines in one facility with matched housing temperature and diet, reading UCP1, core temperature and indirect calorimetry, and stratify by circulating HMW fraction.

Hypothesis: Adiponectin suppresses beta3-adrenergic BAT activation acutely while promoting chronic beiging of subcutaneous fat through T-cadherin on M2 macrophages, so the two results are depot- and timescale-specific rather than contradictory.

Experiment: Test directly whether adiponectin binds sialic acid, the claim GO:0033691 makes. Assay purified recombinant adiponectin against a sialylated glycan array and by surface plasmon resonance against free Neu5Ac and sialyllactose, with a Siglec ectodomain as positive control and desialylated adiponectin as an internal control.

Hypothesis: Adiponectin shows no measurable sialic acid binding; the annotation originates from a post-translational modification of adiponectin itself, and the array will be negative at all concentrations.

Experiment: Establish the human interactome of adiponectin with an assay compatible with its topology. Perform secretome-scale extracellular interaction screening (for example AVEXIS or an equivalent avidity-based assay using eukaryotically expressed, multimerised ectodomain baits) against a library of human cell-surface and secreted proteins, and confirm hits by surface plasmon resonance with size-fractionated trimer, hexamer and HMW adiponectin.

Hypothesis: None of the 59 unreplicated two-hybrid partners will reproduce, whereas AdipoR1, AdipoR2, T-cadherin, CD93 and PDGF-BB will, and several hits will be specific to the hexamer/HMW species.

Experiment: Separate ligand sequestration from receptor signalling in the vascular phenotype. Compare wild-type adiponectin with a C1q-domain point mutant that retains AdipoR binding but loses PDGF-BB binding (identified by alanine scanning guided by the 6U66 structure) for their ability to suppress PDGF-BB-driven proliferation and migration of human aortic smooth muscle cells.

Hypothesis: The anti-proliferative effect depends on direct PDGF-BB sequestration rather than on AdipoR signalling, so the separation-of-function mutant will lose the vascular effect while retaining AMPK activation.

Experiment: Determine whether the podocyte effect is filtration or reabsorption, which decides between GO:2000534 and the proposed glomerular permeability term. Combine micropuncture or FITC-albumin two-photon imaging of the glomerular filtration barrier with megalin/cubilin-dependent proximal tubule uptake assays in adiponectin-treated and untreated Adipoq-null mice.

Hypothesis: Adiponectin reduces trans-barrier albumin flux with no change in tubular reabsorptive capacity, confirming that the existing absorption term names the wrong process.

Knowledge Gaps

What is not known β€” curated, literature-grounded statements of the open unknowns (the inverse of core functions).

Gap: No structure of adiponectin bound to any of its three receptors exists. The three deposited structures (4DOU, 6U66, 6U6N) all cover only the globular C1q domain, residues 104-244, so neither the collagen-like domain nor any receptor complex has been resolved, and the structural basis of multimer-selective T-cadherin binding is unknown.

OPEN BIOLOGY RESIDUAL_SUBGAP

What is known: The globular C1q domain fold is solved to 0.99 A (PDB 6U66, residues 107-244), the trimer/hexamer/HMW ladder and its disulfide dependence are established, and the three receptors are identified.

Significance: Multimer-selective receptor engagement is the central unexplained feature of adiponectin biology and the main obstacle to designing receptor-selective agonists.

Provenance (the field's own admissions):

Gap: Whether the globular fragment of adiponectin, which is far more potent than the full-length protein in many assays, is actually generated in vivo in humans remains unresolved. It can be released from full-length adiponectin by neutrophil elastase in vitro, but whether it circulates at meaningful concentrations is disputed, and much of the receptor literature uses it as a surrogate.

OPEN BIOLOGY RESIDUAL_SUBGAP

What is known: Neutrophil elastase can release the globular fragment from full-length adiponectin in vitro, and the fragment binds AdipoR1/AdipoR2 and activates AMPK; what is unresolved is whether it exists in human circulation at functionally relevant levels.

Significance: A large fraction of the functional literature, and therefore of this gene's GO record, rests on globular adiponectin whose physiological existence is uncertain.

Provenance (the field's own admissions):

Gap: The sign of adiponectin's effect on inflammation is unresolved and appears to depend on multimer state and target-cell polarisation. HMW adiponectin induces MCP-1 and IL-8 in peripheral blood mononuclear cells and microvascular endothelium while LMW does not, and hexamer/HMW species activate NF-kB in myocytes, yet adiponectin suppresses TNF-induced NF-kB in endothelium. GOA carries both directions with no marker of the conditions that distinguish them.

OPEN BIOLOGYONTOLOGY RESIDUAL_SUBGAP

What is known: The multimer species are separable and individually assayable, and their divergent effects on AMPK, NF-kB and T-cadherin binding are reproducible; what is missing is a rule predicting the sign in a given cell type, and any way to record multimer state in a GO annotation.

Significance: Both GO:0043123 and GO:0043124, and both GO:0050728 and the chemokine-induction rows, are simultaneously annotated; without a convention for recording multimer state the record cannot express what is actually known.

Provenance (the field's own admissions):

Deep Research

Affinage

(ADIPOQ-deep-research-affinage.md)

Loading supporting content…

Download this section (compressed HTML)

πŸ“š Additional Documentation

Notes

(ADIPOQ-notes.md)

Loading supporting content…

Download this section (compressed HTML)

Bioinformatics Results

(RESULTS.md)

Loading supporting content…

Download this section (compressed HTML)

πŸ“„ View Raw YAML

Loading supporting content…

Download this section (compressed HTML)