AADACL2 (Q6P093) — review notes

Human arylacetamide deacetylase-like 2, HGNC:24427, chromosome 3, 401 aa, PE 1: Evidence at protein level, Pharos Tdark. Reviewed 2026-07-25 for the PAINT + affinage campaign.

What the record contains

Seven GOA rows, no experimental evidence of any kind: 2 IBA (PANTHER) and 5 IEA
(InterPro ×2, ARBA ×2, UniProt SubCell ×1). Three rows are activities at three levels of
generality; three are locations, two of which are incompatible readings of the same
N-terminal helix.

Literature: genuinely absent for function, present for genetics

PubMed returns 6 hits for AADACL2; none characterises the protein. The affinage record came
back empty (n_discoveries: 0, citation_count: 0, no self_evaluation_pairwise score), and
in this case
the emptiness is corroborated rather than taken on trust — the only literature statement
about AADACL2's function is that there isn't one:

• PMID:35736449
• PMID:35736449

The three UniProt RN references are all large-scale sequencing (PMID:17974005 German cDNA
consortium, which supplied isoform 2; PMID:16641997 chromosome 3; PMID:15489334 MGC, which
supplied the Ser-186 variant). None assays the protein. Note the campaign rule: PE 1 means
the protein is detected, so the honest statement is that there are no
functional/biochemical data, not that there are no data.

Two genetic associations exist and neither supports a GO annotation:

• Psoriasis exome-indel screen in 32,043 Chinese Han individuals:
  PMID:31078570
• SIREN ischemic-stroke GWAS in indigenous Africans:
  PMID:38317187, with
  PMID:38317187

The stroke signal is explicitly near the gene, not in it. A skin-expressed gene surfacing
in a psoriasis screen is at least tissue-coherent; a locus association is not a function.

Family context

AADAC is the characterised member: a liver/intestine enzyme that hydrolyses both amide bonds
of arylacetamide xenobiotics and ester bonds of diglycerides, and it is
PMID:35736449.
AADACL1/NCEH1/KIAA1363 is the other characterised member and is likewise membrane-anchored:
PMID:35736449-long
hydrophobic stretch. The review also notes
PMID:35736449
(hormone-sensitive lipase) — the GDXG / HSL-like α/β-hydrolase clan that AADACL2 belongs to
per [file:human/AADACL2/AADACL2-uniprot.txt "Belongs to the 'GDXG' lipolytic enzyme family."].

Expression: HPA reports AADACL2 as skin-restricted (Tissue enriched (skin),
"detected in single", 10.1 nTPM in skin) — the most tissue-restricted member of the family
after AADACL4. AADACL3 is also skin/placenta-expressed, so the paralogous cluster looks like a
skin-associated expansion of the AADAC family.

Three things worth curating

1. The activity is real machinery, not a fold-derived label

The campaign's standing warning is that a domain's name becomes an activity in GO. The
mirror-image error is to dismiss such a call without checking the residues. Here the residues
are present:
[file:human/AADACL2/AADACL2-uniprot.txt "FT ACT_SITE 189"], 341 and 371 (Ser-Asp-His),
an oxyanion-hole motif at
[file:human/AADACL2/AADACL2-uniprot.txt "FT MOTIF 111..113"], a GDSSG nucleophile
elbow at 187-191, a conserved 116-338 disulfide, and a PROSITE PS01174 GDXG-lipase-serine
match. The reciprocal alignment test in
[file:human/AADACL2/AADACL2-bioinformatics/RESULTS.md "Eight of the fifteen sources with annotated active sites map all three residues onto"]
AADACL2's own annotated triad. Those eight are the seven closest sources plus M. tuberculosis
LipN at 26.1% identity, and every source that fails lies at 26.5% identity or below, where the
global alignment loses register in the C-terminal half. They include three independently
characterised M. tuberculosis carboxylesterases and the 51.6%-identical human paralog AADAC.

That licenses GO:0017171 serine hydrolase activity, whose definition is exactly a serine
nucleophile activated by an acid/base proton relay. GOA currently gives AADACL2 only
GO:0016787 hydrolase activity and GO:0003824 catalytic activity from the mechanism side,
which is a strictly weaker statement than the protein's own feature table supports.

2. The PANTHER node placement is inverted, and it affects three genes

From RESULTS.md section 5: GO:0017171 sits only at the ortholog-specific nodes
PTN002745055 (AADAC) and PTN002745068 (NCEH1), so AADACL2/3/4 inherit no mechanism term.
What they do inherit at the shared family node PTN009058713 is GO:0016020 membrane.
This is backwards with respect to what transfers safely: the catalytic triad is conserved in
all five human members; the N-terminal anchor is not (12 of the 14 reviewed IPR017157
entries carry a TRANSMEM, 11 of them an explicit type-II signal anchor, and AADACL2 is the
only one whose equivalent segment is annotated as a cleaved signal peptide instead). Moving GO:0017171 down to
PTN009058713 and dropping GO:0016020 from it would correct AADACL2, AADACL3 and AADACL4
in one edit.

3. Where I stopped short: the localisation is genuinely unresolved

My first read was "membrane is a paralog artefact; secreted is right". Measurement did not
support that, and this is recorded here so the reasoning is auditable:

• UniProt's cleavage site (after residue 18) puts His at -1, which signal peptidase I
  essentially does not accept (-1 is small/neutral A/S/C/G/T in the great majority of sites).
• The hydrophobic core is more hydrophobic than AADAC's experimentally confirmed type-II
  signal anchor (peak KD-19 mean 1.94 vs 1.81), and the two N-termini are homologous over the
  whole segment with the same downstream TP.PDN.EE.W motif. Hydropathy cannot tell a
  cleaved signal peptide from a signal anchor here.
• HPA's predictor vote does put AADACL2 (and only AADACL2) in the secreted class, but the same
  HPA pipeline fails to classify AADAC as a membrane protein, so it cannot be used as
  independent corroboration.
• All four observed peptides lie in the catalytic domain (earliest residue 124). Mass
  spectrometry establishes expression and says nothing about the N-terminal fate.

So both GO:0005576 and GO:0016020 are unsupported computational readings of the same
segment, and neither can be accepted or removed on present evidence. Actions: UNDECIDED for
all three location rows, with the experiment that would settle it recorded in
suggested_experiments. Recording the contradiction is the deliverable here; picking a
winner would have been a guess dressed as a finding.

Isoform note

Isoform 2 (Q6P093-3) replaces residues 47-53 and deletes 54-401, and UniProt flags it as
likely NMD-degraded:
[file:human/AADACL2/AADACL2-uniprot.txt "premature stop codon in the mRNA, leading to nonsense-mediated mRNA"]
decay. It cannot be catalytically active (it stops before Ser189), so no annotation should be
attributed to it. All annotations here concern isoform 1 (Q6P093-1), the MANE-Select product.

Curiosity, not evidence

The UniProt record carries two DrugBank cross-references, DB07814 Gibberellic acid and
DB07815 Gibberellin A4. The plant gibberellin receptor GID1 is a derivative of this same
GDXG/HSL-like carboxylesterase clan, so these are almost certainly structure-similarity
mappings rather than any claim about human ligands. Not used for anything in the review.

Round-1 review follow-ups (PR #2266, approved with five non-blocking suggestions)

All five were taken:

1. The aggregate 8/15 triad figure understated the nucleophile. results.json already had the
   per-position data: all 15 sources align their own catalytic serine onto position 189, and
   14 of 15 carry a Ser there (the exception is soybean HIDH, whose own nucleophile is Thr —
   and HIDH is the one source that is a dehydratase, not a hydrolase). The 8/15 figure is
   driven entirely by the acid and base drifting out of register below 26.5% identity. Added as
   a table in RESULTS.md section 1, computed by the script (per_target_active_site), and
   surfaced in the GO:0017171 MODIFY reason.
2. The GO:0052689 ACCEPT reason called that term "the best available statement of the core
   function" while core_functions uses GO:0017171. Reconciled: GO:0052689 is the most
   informative activity statement on the record, but it is inferred from family membership,
   whereas the serine-hydrolase mechanism is evidenced on this protein's own residues and is
   also what survives if AADACL2 prefers amide over ester bonds, as AADAC does.
3. "Mutually exclusive" overstated GO semantics — GO:0005576 and GO:0016020 are not
   disjoint (a shed ectodomain can be in both). Narrowed everywhere to what is actually
   claimed: two incompatible readings of one hydrophobic helix, only one of which can describe
   the mature protein.
4. The InterPro2GO GO:0016787 MODIFY now says where the fix lands — the
   IPR013094→GO:0016787 mapping should not change, since a bare alpha/beta-hydrolase-3 match
   warrants nothing more; the correction is a protein-level annotation on AADACL2.
5. Added PMID:16641997 (chromosome 3 sequencing, UniProt RN[2]) to references: so the
   "all three UniProt references are large-scale sequencing" claim is self-contained.

Items 2 and 4 above are superseded by the cross-gene adjudication section at the end of
this file. The reconciliation in item 2 was the wrong way round — GO:0052689 is now the
recorded core function — and the "protein-level annotation on AADACL2" framing in item 4
proposed a term (GO:0017171) that one of the row's own donors refutes.

Actions taken

(The three MODIFY rows originally proposed GO:0017171 serine hydrolase activity; see the
cross-gene adjudication below for why they now all collapse onto GO:0052689 instead.)

Cross-gene adjudication of the PTN009058710 GO:0016787 row (AADACL2 / AADACL3 / AADACL4)

The defect. AADACL2, AADACL3 and AADACL4 each carry one GO:0016787 hydrolase activity
IBA row from GO_REF:0000033, transferred from PANTHER node PTN009058710, and the
WITH/FROM fields are byte-identical across the three records — the same 17 tokens. Three
separate reviews nevertheless reached three different verdicts on that one row:

How it was settled. By measurement, not by preference. The shared node audit
AADACL2-bioinformatics/audit_node_PTN009058710.py resolves all 17 tokens (16 proteins plus
the tree node itself) and reads each donor's chemistry off its own EC numbers and its own
curated GO annotations classified by fetched ontology ancestry, and its nucleophile off its own
ACT_SITE features. Results in NODE_PTN009058710.md:

GO:0016787 hydrolase activity:                  TRUE 16, FALSE 0, UNDETERMINED 0
GO:0052689 carboxylic ester hydrolase activity: TRUE 14, FALSE 2, UNDETERMINED 0
GO:0017171 serine hydrolase activity:           TRUE 15, FALSE 1, UNDETERMINED 0

Neither refinement is true of the whole node, and the two refutations lie on different
axes, so neither can be rescued by choosing the other:

• the bond-type axis is blocked by the two kynurenine formamidases, mouse Afmid (Q8K4H1,
  GO:0004061 by IMP) and yeast BNA7 (Q04066, GO:0004061 by IDA), both EC 3.5.1.9. Their
  term sits under GO:0016810 (C–N bonds), a sibling of the ester branch.
• the mechanism axis is blocked by soybean HIDH (Q5NUF3), whose nucleophile-elbow residue is
  Thr164, against GO:0017171's definition demanding "a catalytic triad consisting of a
  serine nucleophile".

Because GO:0016788 and GO:0016810 are siblings whose only common ancestor below
GO:0003824 is GO:0016787, the term PAINT chose is the exact LCA of its donor set.
So AADACL3's reading was the correct one, and all three genes are now harmonised to it:
MODIFY → GO:0052689 on redundancy grounds only, root_cause: EVIDENCE_CIRCULAR_OR_REDUNDANT, and no failure_modes. GRANULARITY_MISMATCH is dropped
because it presupposes donors that agree with a term still sitting above them; here the donors
disagree and the parent is their LCA, so there is no granularity defect at all.

Two premises that were wrong and mattered.

1. Both merged reviews leaned on soybean HIDH being "not a hydrolase at all but a dehydratase",
   which would have threatened even GO:0016787. It is bifunctional: GO:0033987 dehydratase
   by IDA and GO:0106435 carboxylesterase activity by IDA, EC 4.2.1.105 and EC
   3.1.1.1. So it refutes the serine term only, and the ester term is blocked by the two
   formamidases instead. This sentence in RESULTS.md was corrected.
2. Yeast BNA7 does resolve — xref:sgd-S000002836 → Q04066, ACT_SITE 110 labelled
   "Nucleophile" by UniProt and reading as Ser. The AADACL3 audit reached it through an
   Alliance record and reported the nucleophile unresolved, undercounting the serine tally;
   with BNA7 in, it is 15 of 16 serine, one threonine.

Where the mechanism term does belong. GO:0017171 is not wrong about this family, it is
attached to the wrong node. At the family node PTN009058713 — whose WITH/FROM names only
human AADAC, mouse Aadac and mouse Nceh1 — the term is true of every donor PAINT cites there
and held by IDA in all three of them, and all three are IPR017157 members. All three blockers at the deep
node lie outside IPR017157. So the PAINT recommendation is a node move
(PTN002745055/PTN002745068 → PTN009058713), not a term change on the row, and it is now
stated that way in knowledge_gaps and suggested_questions in AADACL2 and in
suggested_questions in AADACL4.

Knock-on change to core_functions. AADACL2's core molecular function moves from
GO:0017171 to GO:0052689, matching AADACL4 and AADACL3, which have identical catalytic
registers and the same subfamily signature. GO:0017171 remains true of AADACL2 on its own
residues and is stated as such in the core_functions description, but no existing row carries
it and it is pursued as the node move above rather than asserted as a second activity. The
GO:0003824 ARBA row's replacement was moved to GO:0052689 for the same reason: all three
general molecular-function rows on this record now collapse onto the one specific term the
record actually has.

Round-2 corrections to the adjudication (PR #2286 review)

Five factual/rhetorical items, all conceded after checking:

1. The three-gene equality was asserted but only measured for two. genes/human/AADACL3/ is
   not in the tree while #2264 is open, so the first version of the audit recorded
   genes_sharing_the_row: ["AADACL2","AADACL4"] while the prose claimed all three. Fixed by
   measuring rather than softening: the audit now fetches the audited row's WITH/FROM set per
   accession from QuickGO for Q6P093, Q5VUY0 and Q5VUY2, cross-checks it against the committed
   TSVs wherever both exist, and treats a gene covered by neither source as a hard error. AADACL3's
   set comes back as the same 17 tokens, so the claim now rests on measurement for all three.
   The script's missing-TSV path also prints the absence and its reason instead of returning a
   silent {"present": False}, which is what its own docstring had promised.
2. The family node is IDA-supported by all three donors, not two. Human AADAC (IBA,IDA,IEA),
   mouse Aadac (IBA,IDA,IEA,ISO) and mouse Nceh1 (IBA,IDA) all hold GO:0017171
   experimentally. The audit now records this per member (family_node.mechanism_term_support), so
   the node-move recommendation is stronger than first stated.
3. The Thr164 call is a fold-position inference and the whole verdict turns on it. UniProt
   describes HIDH's ACT_SITE 164 as "Proton acceptor" (ECO:0000305), not as a nucleophile —
   unlike Afmid and BNA7, where the nucleophile is labelled. Two things now carry the call
   instead of the label: the elbow pentapeptide, added to the audit for every donor, shows 15 of
   16 reading G-x-S-x-G (GDSAG ×12, GQSAG, GHSAG, GHSVG) while HIDH alone reads GETSG;
   and a sensitivity analysis shows the verdict is robust — downgrading HIDH to undetermined
   still leaves GO:0017171 untrue of every donor, and downgrading every positionally-inferred
   nucleophile leaves it supportable for only the 2 donors UniProt labels explicitly.
4. "All fifteen donors place their catalytic serine on position 189" contradicted the threonine
   donor. It is the catalytic nucleophile that aligns on 189 in 15 of 15, with a serine there in
   14 of 15. Corrected in the row summary and in core_functions.
5. The core_functions rationale leaned on cross-gene consistency. It now leads with the
   evidence that actually carries GO:0052689 — IPR017157 is a family-specific signature over
   almost the whole chain, and every biochemically characterised family member is a carboxylic ester
   hydrolase — with the paralog consistency noted afterwards rather than doing the work.

One item held, with the reasoning recorded. The reviewer notes that dropping
GRANULARITY_MISMATCH goes beyond the enum's literal definition ("parent term is true but
uninformative"), which would fit this row. Held, because failure_modes is documented as the
biological shape of a propagation issue and this propagation has none: the parent is
uninformative because the donors are heterogeneous, not because the transfer could have been more
specific. That is the rule this campaign already adopted ("ask whether the term is the LCA of its
donors; GRANULARITY_MISMATCH is only apt when the donors agree and the term still sits above
them"). The reasoning is now stated inline in both hydrolase rows rather than left implicit, and
the schema's enum description is flagged as worth clarifying so the two readings stop being
interchangeable.

Round-3 tidy-up (the five optional items from the #2286 review)

All five taken; none changed a conclusion.

1. The GRANULARITY_MISMATCH argument was stated twice in the same reason field — the round-1
   sentence survived alongside the fuller round-2 paragraph. Round-1 sentence dropped, and the
   IPR013094 IEA row now cross-references the argument on the IBA row instead of repeating the
   whole paragraph verbatim.
2. RESULTS.md still carried the self-contradicting sentence the YAML had already fixed ("places
   its own catalytic serine on position 189, and fourteen of them carry a serine there").
   Corrected at source to "catalytic nucleophile", and the supporting_text that quotes it
   moved with it — the fix had to be made in both places or the quote would have gone stale.
3. "True of every donor at the family node" was an inference stated as a measurement: the
   family donor set is the WITH/FROM of the GO:0016020 row, and PTN009058713's membership is
   never enumerated because the PANTHER tree is not fetched. Reworded throughout to "every donor
   PAINT cites at that node", with membership_enumerated: false recorded in the audit JSON.
   The independently measured half of the argument is the IPR017157 split, which is per donor.
4. Two script infelicities: the uncovered guard could not fire (from_quickgo is built from
   PARALOG_ACCESSIONS, which covers all three genes), so it is now an assertion documenting the
   invariant — it fires if a gene is ever added to GENES without an accession; and
   family_mechanism silently skipped an unresolvable family-row token, which now dies, matching
   what query_for already refuses to do.
5. Two supporting_text entries quoted the same file with one a strict prefix of the other, the
   longer starting mid-bold. Reduced to the single clean quote.

Round-4: the same staleness, in the file that flagged it

The reviewer found that AADACL2-bioinformatics/RESULTS.md still carried "true of every donor at
PTN009058713 and IDA-supported by two of them" — a sentence this PR itself added, left
un-updated when round 2 changed "two" to "three" in six other places and round 3 changed "every
donor at" to "every donor PAINT cites at". So both conceded items were unapplied in one
sentence, in the very file whose stale-quote problem was item 9. Patching named line numbers is
what let it survive twice.

Fixed by sweeping instead of patching: a single script now greps all seven changed files for
every IDA … two of them variant and every unqualified every donor at the family node, applies
the corrections idempotently, and then re-greps and hard-fails if any occurrence survives.
That found four more instances beyond the one reported — the round-1 sentence in both notes
files, the "Not two of three: every donor at the family node" line in NODE_PTN009058710.md, and
two places in the AADACL4 review that round 3's line-targeted edit had missed. Lesson for the
campaign log: when a phrase is corrected, grep the whole changed file set for the phrase, not for
the line.

Also from the same review: the uncovered guard is now an explicit if … die() rather than an
assert, since python -O strips assertions and the guard is the mechanism by which the
"no gene drops out of the equality test" promise is kept.

Round-5: retiring the divergence statements this harmonisation made false

Extending the harmonisation to AADACL3 invalidated every statement in the three reviews that
described them as disagreeing — and the round-4 sweep grepped for the two of them phrasings but
not for those, which is how they survived a round.

• AADACL3's suggested_questions said the merged AADACL2 review resolves the same row as
  MODIFY → GO:0017171 with TERM_SCOPING_PROBLEM + GRANULARITY_MISMATCH, that "both cannot be
  right about one row", and that "PR #2266 needs a follow-up to settle it". None of that
  survives: AADACL2 now carries AADACL3's own verdict. Rewritten to ask PAINT only where the
  mechanism term should sit — the question that is genuinely still open — while keeping the
  merges-two-existing-rows argument, which is what settled the matter.
• AADACL2's and AADACL4's row reasons motivated the schema request with "since AADACL3's review
  reaches the same LCA conclusion while keeping the mode on the literal reading". That divergence
  no longer exists either. The request stands but the justification changes: all three paralogs now
  encode the row on the propagation-shape reading, which the enum text does not itself state, so
  the convention is carried by argument rather than by the schema.
• The shared audit's "Why this audit exists" section is explicitly marked historical, with a
  closing "Settled outcome" paragraph, so the one remaining mention of TERM_SCOPING_PROBLEM
  cannot be read as a live claim.
• AADACL3's earlier conflict section (in AADACL3-notes.md) is rewritten in the past tense
  with the resolution stated up front, rather than opening "This review now conflicts with the
  merged AADACL2 review … a curator reading both files today gets contradictory advice".
• AADACL3's 13-of-14 serine count now carries the pointer to the shared audit's 15 of 16 at
  the place the count is stated (AADACL3-ai-review.yaml, the analysis review_notes), not only
  in a propagation_review comment and a reference entry elsewhere in the file.

The automation, which is the actual fix

Five items on this PR (9, 13, 17, 18, 22) were the same defect: a claim corrected in one place and
left standing in another, twice in a file that recorded the lesson. Round 5's own bullet asserted
two of these fixes that the tree did not contain. Being more careful demonstrably does not work, so
the checks are now a committed script:

genes/human/AADACL2/AADACL2-bioinformatics/check_paralog_agreement.py

It enforces two things across AADACL2, AADACL3 and AADACL4 — the reviews, the notes files and
the audit prose:

1. the agreement invariant — for each of the six GO:0016787 rows: MODIFY →
   GO:0052689, root_cause: EVIDENCE_CIRCULAR_OR_REDUNDANT, no GRANULARITY_MISMATCH,
   supporting_entities equal to that gene's own GOA WITH/FROM column, one shared 17-token set
   across all three genes, core_functions molecular function GO:0052689, and the shared audit
   cited;
2. the stale-claim greps — no live TERM_SCOPING_PROBLEM, "both cannot be right", "needs a
   follow-up", IDA … two of them, unqualified "every donor at the family node", "13 of 14"
   without a superseding pointer, or "not yet in the tree", in any of those files. Text
   explicitly marked historical is exempt, and the exemption is itself checked.

All eleven guards were verified by deliberately breaking them: --self-test copies the
tree to a temporary directory, applies one mutation at a time, and requires each to be caught. That
paid for itself immediately — the first run reported superseding pointer removed from a count: NOT caught, and the cause was the mutation, which only reworded a lead-in and left the pointer
inside the search window, so nothing was actually broken and the guard was right to stay silent.
Reading the guard would not have found that; only trying to break it did. The mutation now deletes
the whole clause and raises if its target text has moved, so the self-test cannot silently pass
later.

Two design points worth recording. Curator-facing text (reviews, the audit prose, RESULTS.md) is
grepped wholesale, but notes files are journals — a journal recording "X was wrong, now fixed"
necessarily contains X, so a blanket grep is unusable there. They are scanned paragraph by
paragraph and a stale phrase is allowed only where the paragraph, or a marker at the top of its
section, marks the passage retrospective; an unqualified stale sentence in running prose fails,
which is exactly the shape of the AADACL3 section that survived four rounds. And the section-level
exemption requires a strong marker (superseded, historical, since resolved …) rather than any
past tense, so appending a new live claim to an old section is not laundered by its header — there
is a self-test mutation for precisely that.

Integration with just is out of scope for a gene PR, so it runs as
uv run --no-project --with pyyaml python check_paralog_agreement.py and is documented in the
audit file.

Generalisable lesson, and the reason this is a script rather than a resolution: when a change
makes a claim false, grep for the claim, not for the sentence you remember writing — and when the
change is "these two now agree", the claims to hunt are the ones asserting that they do not.