What actually places ABRACL in the cilium?

Generated by analyze_cilium_evidence.py. All numbers are fetched live from the IntAct and UniProt REST APIs at run time; nothing here is hard-coded.

Question

Human ABRACL carries GO:0005929 cilium (IEA, GO_REF:0000044), mapped from the UniProt subcellular-location line Cell projection, cilium {ECO:0000305|PubMed:37759737}. That publication is an expression study of mouse and cat embryonic telencephalon which does not assay ciliary localisation; it attributes the claim to a high-throughput proteomics study. This asks what that primary evidence is, and whether a ciliary role is plausible for the family.

Part 1 - the physical evidence, from IntAct

Interactions recorded for Q9P1F3 5
Distinct detection methods 1
Distinct source publications 1
Partner Detection method Interaction type Publication
CNTRL proximity-dependent biotin identification proximity 26638075 (pubmed)
CNTRL proximity-dependent biotin identification proximity 26638075 (pubmed)
DCTN1 proximity-dependent biotin identification proximity 26638075 (pubmed)
RPGRIP1L proximity-dependent biotin identification proximity 26638075 (pubmed)
SASS6 proximity-dependent biotin identification proximity 26638075 (pubmed)

Detection methods observed: proximity-dependent biotin identification.

Part 2 - is the family present where cilia do not exist?

Counts are UniProtKB entries (reviewed + unreviewed) cross-referenced to each PANTHER family in each proteome. Angiosperms build no cilium, basal body or centriole at any life stage, and neither does Dictyostelium discoideum - which is also the donor organism for ABRACL's IBA annotation. A family whose conserved function is ciliary should score zero in all four.

PANTHER Family Role Homo sapiens Danio rerio Chlamydomonas reinhardtii Arabidopsis thaliana Oryza sativa subsp. japonica Zea mays Dictyostelium discoideum
PTHR46334 Costars / ABRACL test 1 1 0 2 1 3 1
PTHR44117 IFT88 cilium-core control 7 8 2 0 0 0 0
PTHR12969 IFT52 cilium-core control 2 4 3 0 0 0 0
PTHR20870 BBS1 cilium-core control 11 3 1 0 0 0 0
PTHR46090 ARL13B cilium-core control 18 13 2 0 0 0 0
PTHR11588 tubulin beta universal control 122 54 11 23 22 98 4

Read-out

Interpretation

Part 1 shows the entire physical-interaction record for ABRACL is one proximity-labelling experiment. BioID biotinylates whatever comes within roughly 10 nm of the bait during hours of labelling, so a prey list is a neighbourhood, not a localisation call; and three of the four baits here (SASS6, CNTRL, DCTN1) are centriolar or dynactin proteins rather than ciliary ones. Part 2 shows the family is retained in four proteomes that build no cilium at all, while four cilium-core families are absent from every one of them and the universal control is present in all. The reciprocal observation points the same way: no PTHR46334 member is cross-referenced in Chlamydomonas reinhardtii, the most heavily ciliated proteome in the panel, where all four cilium-core controls are present. Costars distribution therefore does not track cilium presence in either direction. That single absence is a cross-reference count rather than a homology search, so it is reported as an observation and no weight is placed on it alone.

Neither part proves ABRACL is absent from human cilia; a lineage-specific ciliary role would not show up in a phyletic comparison, and BioID hits are sometimes real. What they establish is narrower and sufficient for curation: there is no direct localisation evidence behind the cilium annotation, and the conserved function of the family cannot be ciliary.