This manual source account distinguishes selected original experiments from abstract-only caches and inferred mechanisms. It is not an automated provider report. Reference assessments in the review give the access scope for each source.
PMID:11714716(https://www.med.upenn.edu/dreyfusslab/publications/2002GubitzJBC5691.pdf) used full-length cellular GEMIN5 and a bacterial 43–1508 construct. Purified SMN and SmB interactions are distinct from reticulocyte-translation and cellular association assays. The six-page main text and captions were read; Figures 3–5 on page 5634 were inspected. The 694–1508 immunogen is an experimental reagent.
PMID:16857593(https://pubmed.ncbi.nlm.nih.gov/16857593/) supports snRNA recognition and Sm-core assembly in human-cell experiments. Reduced assembly after depletion is not a direct spliceosomal splicing assay. PMID:20513430(https://pmc.ncbi.nlm.nih.gov/articles/PMC2901871/) extends the substrate-delivery model to pre-snRNAs. Selected original Methods and Results were read for both.
The original PMID:19377484 article was accessed as an author-article transcription, with the official supplement separately inspected. Figure 5a reports positive binding to recombinant SMN–Gemin2. The 1–720 and 1–844 fragments differ in U4 versus U1 recognition; they are not natural isoforms. Native subcomplex stoichiometry remains unresolved.
PMID:17178713(https://www.researchgate.net/publication/6620949_A_Comprehensive_Interaction_Map_of_the_Human_Survival_of_Motor_Neuron_SMN_Complex) reports conditional recruitment and interactions in several assay systems. PMID:18984161(https://pure.mpg.de/rest/items/item_641696_5/component/file_641695/content) includes native human SMN complexes containing GEMIN5, while simplified complexes retain assembly activity without it. Selected Methods, Results and captions were read; Figure 3 of the latter was inspected. These observations do not establish universal GEMIN5 dispensability or autonomous assembly catalysis.
PMID:19750007(https://pmc.ncbi.nlm.nih.gov/articles/PMC2736588/) establishes cap binding, but its proposed W286 stacking mechanism was a model. The human 1–740 structure in PMID:27834343(https://pubmed.ncbi.nlm.nih.gov/27834343/) resolves two seven-bladed propellers and distinct RNA and cap sites. W286 is buried; W286A was insoluble in that study. The original main text, supplementary Methods and captions were read, with selected Figure 1/S1G images. The other 2016 structural papers, 27881600 and 27881601, provide independent construct-specific evidence.
The direct bacterial-protein interaction with eIF4E in PMID:16739988(https://pubmed.ncbi.nlm.nih.gov/16739988/) is retained alongside the tagged-293T no-coimmunoprecipitation result in PMID:19750007. Different conditions need not yield the same association. The exact historical GST construct boundary was not independently re-established; cap-resin recruitment alone is not a direct contact map.
PMID:25911097(https://pubmed.ncbi.nlm.nih.gov/25911097/) combines direct SMN-mRNA recognition with transcript-specific translation effects. PMID:27507887(https://pubmed.ncbi.nlm.nih.gov/27507887/) instead measures ribosome interactions and broader translational repression under its tested conditions. Selected original sections were read; elongation stalling remains a mechanistic proposal. PMID:34424823(https://pubmed.ncbi.nlm.nih.gov/34424823/) examines engineered C-terminal fragments and viral IRES recognition, which does not make those fragments natural products.
The normal PMID:19946888(https://pubmed.ncbi.nlm.nih.gov/19946888/) cache contains the journal abstract. Attempts to retrieve three spreadsheet supplements returned HTML access failures. The following positive target rows come from the authentic, separately accessed 2008 Ghosh thesis PDF, not a verified copy of the final journal supplement:
| Source location | Literal protein identifier | Reported log(e) | Fraction/analysis |
|---|---|---|---|
| Table S7, printed page 325, PDF page 344 | ENSP00000285873 | −23.2 | TFE-resistant YTS membrane pellet, RP analysis |
| Table S8, printed page 388, PDF page 407 | ENSP00000285873 | −3.7 | TFE-resistant YTS membrane pellet, SCX analysis |
Both target rows and their table context were visually checked. The current official Q8TEQ6 cross-reference is ENSP00000285873.6, supporting gene identity. These identification scores are not enrichment measurements. Fraction recovery does not establish integral insertion, membrane topology, direct lipid contact or a particular organelle. The thesis reports 2,278 proteins versus 1,843 in the journal abstract, so the datasets are not silently equated. Modern isoform-specific peptide identity was not established.
PMID:33963192(https://pubmed.ncbi.nlm.nih.gov/33963192/) distinguishes L1068P patient-derived neurons from engineered H913R iPSC clones. Assembly-system abundance changes and extract-level assembly defects do not isolate a catalytic activity of GEMIN5. Selected original Results, Methods, Discussion and captions were read; images and supplements were not included in that reading.
The six current Reactome contexts have explicit human Q8TEQ6 participant paths. R-HSA-191763 is snRNP–Snurportin complex formation; cap hypermethylation is R-HSA-191784. The nuclear-envelope location of import event R-HSA-191830 is not assigned to GEMIN5: its input and output target leaves are cytosolic and nucleoplasmic, respectively. TGS1 and methylosome catalytic roles remain separate.
For large-scale RNA-binding and interaction screens, source-level assay evidence is distinguished from exact target-table verification and from orthogonal physiological validation. In particular, the PMID:22658674 target-table subset was not independently inspected. Historical generated research and old quote associations remain provenance, not fresh verification of their claims.
The annotation-level excerpts are taken from these official record sections. The source ledger stores the retained record hash, version, JSON field and every annotation using that class; the excerpts are not repeated here.
| Record | Section | Scope |
|---|---|---|
| DDX20 Q9UHI6 | SIMILARITY; Pfam DEAD/Helicase_C cross-references | DEAD-box enzyme family supports the partner-class refinement, not helicase catalysis by GEMIN5. |
| eIF4E P06730 | SIMILARITY | eIF4E family identity complements the direct interaction assay. |
| GEMIN2 O14893 | FUNCTION | Sm-assembly organization does not establish a protein-folding chaperone class. |
| GEMIN4 P57678 | SUBUNIT | SMN-complex membership supports its assembly role. |
| SmB P14678 | SUBUNIT | An snRNP subunit is distinguished from the complete RNA–protein particle. |
| SmE P62304, SmD1 P62314, SmD2 P62316, SmD3 P62318 | SIMILARITY | Sm-family identity does not itself establish a specific functional binding replacement. |
| SMN Q16637 | FUNCTION; gene list | The assembly-backbone role is distinct from a signaling scaffold; this accession covers SMN1/SMN2. |
For the specific gem refinement, the original PMID:11714716 Figure 2 localization experiments and its abstract identify SMN-colocalizing gems. The broader HPA row does not independently distinguish gems from other nuclear bodies. For the viral substrate, PMID:34424823 Results, Figure 2 and Table 1 identify d5/d5ss binding by engineered RBS1 constructs. This positive RNA-binding result is not evidence for a natural truncated product or for the exact identity of the incoming RNAcentral reagent. The PMID:16739988 Methods and Figures 4–6 separate recombinant direct binding from reticulocyte/cap-resin and cellular experiments.
The official RNAcentral URS0002318D25 record identifies a 462-nucleotide FMDV type-C IRES in taxon 12116. In PMID:23221641(https://pmc.ncbi.nlm.nih.gov/articles/PMC3553989/), the final Introduction paragraph places the recognized IRES domain upstream of the initiator codon. The Constructs and RNA synthesis Methods, Figure 1 caption and Results on domain-5/C-terminal binding distinguish the RNA segments and purified-protein assays from BHK-21 extract photocrosslinking. Those selected original sections were read from NCBI BioC; figure pixels and supplements were not inspected.
Together with the later d5/d5ss experiments in PMID:34424823, this supports mRNA 5′-UTR binding, GO:0048027. It does not establish that the full RNAcentral sequence equals either short reagent, nor does it convert the engineered protein fragments into natural products.