ACTA2 (P62736) — computed analysis

Regenerate with uv run python analyze_acta2.py; both this file and results.json are
rewritten in full, so git diff after a run is the check that nothing here was hand-edited.

1. GO:0005200 structural constituent of cytoskeleton: who still receives it

PAINT asserts GO:0005200 exactly once in PTHR11937, at node PTN000940351 (IBD, 20250805, 10 seeds), and negates it by IRD at 8 descendant nodes. The assertion count is asserted by the script, so a change to the tree breaks the run rather than the argument.

Human GO:0005200 rows in QuickGO: 389 in total, of which 43 are IBA. The IBA rows resolve to 7 PANTHER nodes:

So the set still receiving the term from the actin node PTN000940351 is 10 genes: ACTA1, ACTA2, ACTC1, ACTG2, ACTL10, ACTL9, ACTR10, ACTRT1, ACTRT2, ACTRT3. ACTA2 is IN that set.

Evidence route per actin-family gene that holds the term at all — the distinction that
makes the IBA set and the holds-the-term set two different sets:

The seeds of the assertion, and what evidence each holds for the term it donated

Is any seed of the assertion also inside a clade the same term was negated in?

Yes — 2 of 10. These proteins supply the experimental support that justifies GO:0005200 at the family root, and PAINT then exempts their own clades from the term it justified:

1b. Which node carries which term, and what is each node FOR

Two different questions. The first cannot find a node that gives a gene nothing it should
have; the second can.

GO:0005884 - PANTHER nodes projecting it onto any conventional human actin:

GO:0017022 - PANTHER nodes projecting it onto any conventional human actin:

GO:0033275 - PANTHER nodes projecting it onto any conventional human actin:

GO:0007015 - PANTHER nodes projecting it onto any conventional human actin:

GO:0005576 - PANTHER nodes projecting it onto any conventional human actin:

Reverse direction - the entire human reach of each node that appeared above:

2. Residue tallies at the nucleotide site and the filament protomer interface

• nucleotide_site: PDB 2BTF chain A, ligands ATP, SR, 4.0 Å heavy-atom cutoff → 19 contact residues.
• filament_interface: PDB 6DJO chain C, 4.0 Å heavy-atom cutoff → 38 contact residues.

Reproduction check: this script reproduces the committed ACTL8 filament-interface tally for 9 shared panel members (ACTL8 (human actin-like 8); ACTB (human beta-actin; IBA donor); ACTG1 (human gamma-actin; IBA donor); ACTA1 (human alpha-skeletal actin; IBA donor); ACTC1 (human alpha-cardiac actin; IBA donor); ACTR2 (human Arp2); ACTR3 (human Arp3); ACTRT1 (human actin-related protein T1); Arp53D (Drosophila actin-like 53D; IBA donor)); a mismatch aborts the run.

Reference-length guard: panel median 376 aa, so anything below 338.4 aa cannot be scored as if every structural position were tested. Flagged: ACTL10 (245 aa, 0.65 of median). outside_span is reported as its own column below so an unreached position can never be added to a substitution count again.

nucleotide site (19 positions)

filament interface (38 positions)

Robustness: the same three proteins under a second substitution matrix and gap model.

3. ACTA2 disease variants against those two surfaces

19 FT VARIANT positions parsed from the cached UniProt entry, mapped onto each structure by the same alignment used for the tallies (never by an assumed offset).

The per-surface table below uses the SINGLE-chain contact sets, which is the right basis
for the cross-species panel in section 2 (every protein is scored on the same positions)
but the wrong basis for this question: chain C of the filament model has no i+2 neighbour,
and actin protomer contacts are not symmetric, so a residue reaching only 'upward' is
invisible from it. The derived counts that follow, and the distance table, use the minimum
over every chain. Reading the single-chain table alone would have said no pathogenic
variant touches another protomer; four do.

All-chain derived counts (the headline figures):

Numbering and contact-set controls

P62736 is a 377-residue precursor and the structures use mature actin numbering, so the
offset should be -2. Verified against three sources that state both numbers, and the
contact set is itself checked against the residues PMID:26637293 names as R179's
inter-strand partners. Any failure aborts the run.

Closest approach, because an absence from a 4 Å set is not a finding

Measured over every chain of the filament model, not one: with four protomers no single
chain has both its i-2 and its i+2 neighbour, so a per-chain answer would silently omit
one strand relationship.

4b. GO:0005515 partners, resolved, and how the interactions were detected

IntAct holds 320 interactions for P62736. Detection methods:

172 of 239 IntAct partners are logged exactly once.

Retraction / erratum / correction status of the PMIDs this review leans on

21 checked, 1 flagged. Both halves are read: the publication-type list AND the cited article's own CommentsCorrections block, because a Publisher Correction is invisible to a publication-type query.

4. WITH/FROM resolution and donor evidence

32 non-experimental GOA rows carry 69 WITH/FROM tokens in total. Counts are derived from the GOA field, with an assertion, because hand-maintained source lists drifted on three genes in this campaign.

GO:0005576 extracellular region (cellular_component, IBA, is_active_in, GO_REF:0000033, assigned by GO_Central) — 2 token(s)

GO:0015629 actin cytoskeleton (cellular_component, IBA, is_active_in, GO_REF:0000033, assigned by GO_Central) — 25 token(s)

GO:0005200 structural constituent of cytoskeleton (molecular_function, IBA, enables, GO_REF:0000033, assigned by GO_Central) — 11 token(s)

GO:0005856 cytoskeleton (cellular_component, IEA, located_in, GO_REF:0000044, assigned by UniProt) — 1 token(s)

GO:0007010 cytoskeleton organization (biological_process, IEA, involved_in, GO_REF:0000108, assigned by GOC) — 1 token(s)

GO:0001725 stress fiber (cellular_component, IEA, located_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0005604 basement membrane (cellular_component, IEA, located_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0006936 muscle contraction (biological_process, IEA, involved_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0015629 actin cytoskeleton (cellular_component, IEA, located_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0019901 protein kinase binding (molecular_function, IEA, enables, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0061870 positive regulation of hepatic stellate cell migration (biological_process, IEA, involved_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0061874 positive regulation of hepatic stellate cell contraction (biological_process, IEA, involved_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0070374 positive regulation of ERK1 and ERK2 cascade (biological_process, IEA, involved_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0071560 cellular response to transforming growth factor beta stimulus (biological_process, IEA, involved_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0072051 juxtaglomerular apparatus development (biological_process, IEA, involved_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:2000491 positive regulation of hepatic stellate cell activation (biological_process, IEA, involved_in, GO_REF:0000107, assigned by Ensembl) — 2 token(s)

GO:0005829 cytosol (cellular_component, TAS, located_in, Reactome:R-HSA-445699, assigned by Reactome) — 0 token(s)

GO:0005829 cytosol (cellular_component, TAS, located_in, Reactome:R-HSA-445700, assigned by Reactome) — 0 token(s)

GO:0005829 cytosol (cellular_component, TAS, located_in, Reactome:R-HSA-445704, assigned by Reactome) — 0 token(s)

GO:0005829 cytosol (cellular_component, TAS, located_in, Reactome:R-HSA-445705, assigned by Reactome) — 0 token(s)

GO:0005829 cytosol (cellular_component, TAS, located_in, Reactome:R-HSA-9604664, assigned by Reactome) — 0 token(s)

GO:0005829 cytosol (cellular_component, TAS, located_in, Reactome:R-HSA-9914537, assigned by Reactome) — 0 token(s)

GO:0005829 cytosol (cellular_component, TAS, located_in, Reactome:R-HSA-9934294, assigned by Reactome) — 0 token(s)

GO:0005829 cytosol (cellular_component, TAS, located_in, Reactome:R-HSA-9934410, assigned by Reactome) — 0 token(s)

GO:0005829 cytosol (cellular_component, TAS, located_in, Reactome:R-HSA-9934486, assigned by Reactome) — 0 token(s)

GO:0016887 ATP hydrolysis activity (molecular_function, ISS, enables, GO_REF:0000024, assigned by UniProt) — 1 token(s)

GO:0010628 positive regulation of gene expression (biological_process, ISS, involved_in, GO_REF:0000024, assigned by AgBase) — 1 token(s)

GO:0030027 lamellipodium (cellular_component, ISS, located_in, GO_REF:0000024, assigned by AgBase) — 1 token(s)

GO:0030175 filopodium (cellular_component, ISS, located_in, GO_REF:0000024, assigned by AgBase) — 1 token(s)

GO:0044297 cell body (cellular_component, ISS, located_in, GO_REF:0000024, assigned by AgBase) — 1 token(s)

GO:0090131 mesenchyme migration (biological_process, ISS, involved_in, GO_REF:0000024, assigned by AgBase) — 1 token(s)

GO:0019901 protein kinase binding (molecular_function, ISS, enables, GO_REF:0000024, assigned by ParkinsonsUK-UCL) — 1 token(s)

5. Reference projection check

For each literature reference on an ACTA2 row: how many annotations does it carry across GOA, and how many DISTINCT entities? A reference that annotates a whole set with identical evidence is one finding projected, not N findings — but only if the phenotype spreads with it, so the per-term entity counts are given too.

• PMID:23533145: unavailable: 1046 annotations exceed the 200-row page this script reads, and a page total is not a whole total; the projection test is unreliable for this reference and is not attempted
• PMID:28514442: unavailable: 3731 annotations exceed the 200-row page this script reads, and a page total is not a whole total; the projection test is unreliable for this reference and is not attempted
• PMID:33961781: unavailable: 9514 annotations exceed the 200-row page this script reads, and a page total is not a whole total; the projection test is unreliable for this reference and is not attempted

PMID:11927518 — per-term entity counts:

PMID:12355421 — per-term entity counts:

PMID:16548883 — per-term entity counts:

PMID:17464107 — per-term entity counts:

PMID:18468998 — per-term entity counts:

PMID:23580065 — per-term entity counts:

Audit