PTHR13723 annotation census for ADAMTSL1

Regenerate with:

uv run genes/human/ADAMTSL1/ADAMTSL1-bioinformatics/check_family_propagation.py

Retrieval timestamp is recorded in results.json so this file stays byte-reproducible.

1. GO:0031012 coverage across human PTHR13723

PAINT holds GO:0031012 as an IBD annotation at node PTN000347317. Of the 26 human members of the family, 24 receive it by IBA from that node.

gene accession subfamily GO:0031012 IBA other rows evidence codes GO:0030198 IBA GO:0030198 other
ADAMTS1 Q9UHI8 PTHR13723:SF40 1 2 HDA, IBA, IEA 1 IBA, IEA
ADAMTS10 Q9H324 PTHR13723:SF26 1 1 IBA, IDA 1 IBA, IEA
ADAMTS12 P58397 PTHR13723:SF189 1 0 IBA 1 IBA, IEA
ADAMTS13 Q76LX8 PTHR13723:SF20 1 1 IBA, TAS 1 IBA, IEA
ADAMTS14 Q8WXS8 PTHR13723:SF24 1 0 IBA 1 IBA, IEA, TAS
ADAMTS15 Q8TE58 PTHR13723:SF39 1 1 IBA, IEA 1 IBA, IEA, ISS
ADAMTS16 Q8TE57 PTHR13723:SF140 1 0 IBA 1 IBA, IEA
ADAMTS17 Q8TE56 PTHR13723:SF151 1 1 IBA, IDA 1 IBA, IEA
ADAMTS18 Q8TE60 PTHR13723:SF167 1 0 IBA 1 IBA, IEA
ADAMTS19 Q8TE59 PTHR13723:SF197 1 0 IBA 1 IBA, IEA, ISS
ADAMTS2 O95450 PTHR13723:SF141 1 1 IBA, IEA 1 IBA, IEA, TAS
ADAMTS20 P59510 PTHR13723:SF165 1 1 IBA, TAS 1 IBA, IEA
ADAMTS3 O15072 PTHR13723:SF158 1 2 IBA, NAS, TAS 1 IBA, IC, IEA, NAS, TAS
ADAMTS4 O75173 PTHR13723:SF38 1 1 HDA, IBA 1 IBA, IEA, TAS
ADAMTS5 Q9UNA0 PTHR13723:SF37 1 2 IBA, IEA, TAS 1 IBA, IEA, ISS, TAS
ADAMTS6 Q9UKP5 PTHR13723:SF27 1 0 IBA 1 IBA, IEA
ADAMTS7 Q9UKP4 PTHR13723:SF142 1 1 IBA, IEA 1 IBA, IEA
ADAMTS8 Q9UP79 PTHR13723:SF41 1 2 IBA, IEA, ISS 1 IBA, IEA
ADAMTS9 Q9P2N4 PTHR13723:SF33 1 2 IBA, IDA, ISS 1 IBA, IEA, ISS
ADAMTSL1 Q8N6G6 PTHR13723:SF157 0 0 - 0 IEA
ADAMTSL2 Q86TH1 PTHR13723:SF147 1 0 IBA 1 IBA, IEA
ADAMTSL3 P82987 PTHR13723:SF169 1 1 IBA, TAS 0 IEA
ADAMTSL4 Q6UY14 PTHR13723:SF144 1 3 HDA, IBA, TAS 1 IBA
ADAMTSL5 Q6ZMM2 PTHR13723:SF173 1 3 IBA, IDA, TAS 0 IEA
PAPLN O95428 PTHR13723:SF281 0 2 IEA, TAS 0 IEA
THSD4 Q6ZMP0 PTHR13723:SF16 1 10 HDA, IBA, IEA, RCA, TAS 1 IBA
Adamtsl1 (mouse) Q8BLI0 PTHR13723:SF157 1 3 HDA, IBA 0 IEA

Members with no GO:0031012 annotation of any kind: ADAMTSL1.

PAPLN holds GO:0031012 without the IBA, but none of its rows is experimental (evidence codes: IEA, TAS), so redundancy suppression does not account for the missing IBA. Its absent IBA is a second coverage gap at this node, not an explained omission.

ADAMTSL1 has neither an IBA nor any other row, and its mouse orthologue - the same PANTHER subfamily, the same IBD node - does receive the IBA.

GO:0030198 coverage, same node

22 of the 26 human members receive GO:0030198 by IBA from PTN000347317: ADAMTS1, ADAMTS10, ADAMTS12, ADAMTS13, ADAMTS14, ADAMTS15, ADAMTS16, ADAMTS17, ADAMTS18, ADAMTS19, ADAMTS2, ADAMTS20, ADAMTS3, ADAMTS4, ADAMTS5, ADAMTS6, ADAMTS7, ADAMTS8, ADAMTS9, ADAMTSL2, ADAMTSL4, THSD4. Members with no GO:0030198 annotation at all: none.

Within the ADAMTS-like branch the IBA reaches only ADAMTSL2, ADAMTSL4, THSD4. ADAMTSL1 and ADAMTSL5 are therefore in the same position for this term - the InterPro IEA and nothing else - which matters when comparing verdicts between their reviews.

2. Hydrolase keyword across the ADAMTS-like branch

UniProt's KW-0378 Hydrolase is what generates the GO:0016787 hydrolase activity cross-reference in an entry's own GO list (IEA:UniProtKB-KW).

gene accession length MF keywords CATALYTIC ACTIVITY comments CAUTION: no metalloprotease domain
ADAMTSL1 Q8N6G6 1762 Hydrolase 0 yes
ADAMTSL2 Q86TH1 951 - 0 yes
ADAMTSL3 P82987 1691 - 0 yes
ADAMTSL4 Q6UY14 1074 - 0 yes
ADAMTSL5 Q6ZMM2 481 Heparin-binding 0 yes
THSD4 Q6ZMP0 1018 Hydrolase 0 no

Hydrolase keyword present on: ADAMTSL1, THSD4. CAUTION stating the metalloprotease and disintegrin-like domains are absent: ADAMTSL1, ADAMTSL2, ADAMTSL3, ADAMTSL4, ADAMTSL5 (absent on THSD4).

Entries with a CATALYTIC ACTIVITY comment: none. So on ADAMTSL1 the keyword sits alongside that entry's own statement that the catalytic domain is missing, with no reaction recorded anywhere in the entry.

3. PAINT's own loss calls in this family

The cached PAINT table records explicitly negated annotations, i.e. terms PAINT blocks from propagating below a node:

node term aspect evidence seed
PTN002673039 GO:0004222 F IKR PANTHER:PTN000347317
PTN002673039 GO:0006508 P IRD PANTHER:PTN000347317

IKR is inferred-from-key-residues and IRD inferred-from-rapid-divergence: PAINT has judged that catalysis was lost on this branch. Where those calls land in GOA:

gene GO:0004222 rows GO:0006508 rows
ADAMTSL1 none none
ADAMTSL2 NOT\|enables IBA GO_REF:0000033 none
ADAMTSL3 none none
ADAMTSL4 none none
ADAMTSL5 none none
THSD4 none none

Negated rows in GOA: ADAMTSL2 GO:0004222. Positive rows in GOA: none. So the loss call is recorded for part of the branch and simply absent for the rest - ADAMTSL1 inherits neither the catalytic terms nor the statement that they do not apply.

4. IBD seed composition at the family node

Counting WITH/FROM tokens on the derived IBA rows overstates the number of experimental sources by one, because GOA appends the PANTHER node itself to the list. The seed lists in the PAINT table give the gene sources directly:

node term aspect evidence seed tokens gene sources by database
PTN000347317 GO:0031012 C IBD 16 16 FB 2, MGI 8, RGD 1, UniProtKB 4, WB 1
PTN000347317 GO:0004222 F IBD 9 9 MGI 3, UniProtKB 6
PTN000347317 GO:0006508 P IBD 13 13 MGI 7, UniProtKB 4, ZFIN 2
PTN000347317 GO:0030198 P IBD 14 14 FB 2, MGI 11, WB 1

Guards

The script aborts rather than emitting a stale sentence if any of these stop holding: a QuickGO response is paginated (page total read as the whole set); human ADAMTSL1 acquires any GO:0031012 row; mouse Adamtsl1 loses its IBA from PTN000347317; ADAMTSL1 loses KW-0378, gains a CATALYTIC ACTIVITY comment, or loses the CAUTION about the missing metalloprotease domain; the IKR loss call disappears from the cached PAINT table; or ADAMTSL1 gains a GO:0004222/GO:0006508 row. A missing cached input is a hard error naming the command that regenerates it. The zero-row count for ADAMTSL1 is produced by the same code path that returns non-zero for the other 25 human members on every run, so the census is its own positive control.