Calcitriol Biosynthesis from Calciol — Obsoletion & Replacement

Bottom line: GO is obsoleting GO:0036378 calcitriol biosynthetic process from calciol and replacing it with GO:1901755 vitamin D3 biosynthetic process, after redefining the replacement to have calcitriol as its primary
output. The change affects 24 curated annotations on five cytochrome P450
enzymes (CYP27B1, CYP2R1, CYP27A1, CYP3A4, CYP24A1) across human, mouse, rat
and zebrafish. We pulled those 24 annotations from QuickGO, matched each enzyme
to its hydroxylation step, and checked the repo for affected reviews, so that
we know which reviews will need refreshing and which unreviewed genes are about
to change. Four of the five enzymes catalyse a vitamin D activation step;
CYP24A1 is an inactivating 24-hydroxylase and is the one case where a
mechanical replaced_by may be wrong. Status: scoped, not started as a project. When the page was
written no affected gene had a review; since then human CYP27A1 has been
reviewed (bile-acid module, #2003), and its GOA file already carries the IDA
row from PMID:15465040 under GO:1901755, which the review keeps as
non-core. The other four genes still have no review here.

Overview

A GO obsoletion proposal will obsolete the biological-process term
GO:0036378 calcitriol biosynthetic process from calciol and replace it
with the sibling term GO:1901755 vitamin D3 biosynthetic process. The
upstream rationale is that the two enzymatic steps captured by GO:0036378
(25-hydroxylation in liver + 1α-hydroxylation in kidney) are the only two
enzymatic steps in the broader MetaCyc vitamin D3 biosynthesis pathway
(MetaCyc:PWY-6076), so the narrower term is redundant once GO:1901755 is
re-defined to have calcitriol (CHEBI:17823, the active hormone) as its
has_primary_output rather than calciol. With that re-definition,
GO:1901755 subsumes the content of GO:0036378 exactly, and a clean
replaced_by is appropriate.

This is a small, well-scoped curation hygiene job: 24 experimental
annotations on 5 cytochrome P450 genes, none of which currently has an
*-ai-review.yaml here. Both groups listed upstream (BHF-UCL, UniProt)
have already marked their portions done in the tracking spreadsheet.

Upstream tickets

• Annotation tracker: geneontology/go-annotation#6449
• Ontology ticket (re-definition + obsoletion):
  geneontology/go-ontology#32077
• Affected-annotation spreadsheet (upstream-curated):
  https://docs.google.com/spreadsheets/d/1L1ymnLg1_t4fNK9psd211WUGOCHBpSd7ny0e1tC9JKo/edit

Obsoletion plan (per upstream)

Term metadata confirmed in OLS on 2026-06-20:

• GO:0036378 (calcitriol biosynthetic process from calciol) — live, slated
  for obsoletion. Synonyms include "vitamin D3 activation" and
  "1alpha,25-dihydroxyvitamin D3 biosynthesis".
• GO:1901755 (vitamin D3 biosynthetic process) — live; the proposed
  replacement target. The upstream ontology ticket records the first two
  proposals (add MetaCyc:PWY-6076 xref; change has_primary_output from
  CHEBI:28940 calciol → CHEBI:17823 calcitriol) as already done; the third
  (obsolete GO:0036378, replaced_by GO:1901755) is still pending.

Affected experimental / curated annotations (24)

Retrieved from the QuickGO annotation API on 2026-06-20
(goId=GO:0036378, manual / experimental evidence codes only). Matches the
upstream group counts after accounting for ZFIN, MGI and RGD assertions that
the upstream tally rolls into "other groups":

On top of these 24 curated records, GO:0036378 currently has ~2,048 total
annotations (QuickGO, 2026-06-20). The non-experimental tail is small for
this term (the first 100 results contained only one IBA + IDA + ~98 IEA),
which is consistent with a tightly-scoped catalytic BP rather than a
broadly-propagated complex/localization term.

Why a clean replaced_by is appropriate here

Unlike GO:0030943 mitochondrion targeting sequence binding
([[MITOCHONDRION_TARGETING_SEQUENCE_BINDING_OBSOLETION]]), where the affected
annotations span biologically distinct mechanism classes, all 24 affected
annotations are on enzymes catalyzing one of the two hydroxylation steps in
the vitamin D3 → calcitriol pathway, and the re-defined GO:1901755
(has_primary_output = calcitriol) literally subsumes the old GO:0036378
content. The four annotated CYP roles are:

• CYP2R1 / CYP3A4 / CYP27A1 — vitamin D 25-hydroxylase activity
  (calciol → 25-hydroxycalciferol / calcidiol). CYP2R1 is the principal
  hepatic 25-hydroxylase; CYP3A4 and CYP27A1 contribute secondary
  25-hydroxylation routes.
• CYP27B1 — 25-hydroxyvitamin D3 1α-hydroxylase
  (calcidiol → calcitriol, in kidney). This is the activating step that the
  obsoleted term explicitly named.
• CYP24A1 — 24-hydroxylase that inactivates 25-hydroxyvitamin D3 and
  1,25-(OH)₂D3 (calcitriol). The IDA annotations for CYP24A1 to GO:0036378
  in PMID:25727742 are arguably borderline: CYP24A1's primary catalytic
  output is the 24-hydroxylated, inactivation-pathway metabolite, not
  calcitriol itself. Reviewers should flag whether GO:1901755 is the right
  successor, or whether a vitamin D catabolic term
  (e.g. GO:0042369 vitamin D catabolic process) is more appropriate.
  This is the one case where mechanical replaced_by may not be the right
  outcome.

Impact on this repo

None of the five affected genes currently has an *-ai-review.yaml under
genes/ (verified 2026-06-20 across genes/human/, genes/mouse/,
genes/rat/, genes/zebrafish/). There are also no other reviews here that
cite GO:0036378 (grep -r "GO:0036378" genes/ returned no hits). So this
obsoletion does not invalidate any existing AI Gene Review content.

That said, vitamin D biosynthesis is a small, well-characterized,
high-textbook-coverage pathway, and CYP27B1 in particular is medically
important (loss-of-function causes vitamin D-dependent rickets type 1A). It
is a natural candidate for a small, focused gene-family review project, with
this obsoletion as a clean entry point.

Scope

• GO branch: Biological Process (single term obsoletion + replaced_by).
• Organisms: Human, mouse, rat (curated experimental set); zebrafish
  (one IDA). No yeast / plant / fly annotations.
• Gene set: 5 cytochrome P450 enzymes — CYP27B1 (1α-hydroxylase),
  CYP2R1 (principal 25-hydroxylase), CYP27A1 (secondary 25-hydroxylase),
  CYP3A4 (alternative 25-hydroxylase), CYP24A1 (24-hydroxylase / inactivator,
  the edge case).
• Type of fix: Curation hygiene + small family review. The biology is
  uncontroversial; the only real curation decision is the CYP24A1 case.

Candidate genes for initial review

Listed in priority order. The first three would directly exercise the
re-defined GO:1901755 once the obsoletion lands.

1. CYP27B1 (human, O15528) — 1α-hydroxylase; the canonical
   "vitamin D activation" enzyme; 9 of the 24 affected annotations. Highest
   priority and best positive control.
2. CYP2R1 (human, Q6VVX0) — principal hepatic 25-hydroxylase; 3 of the
   24 annotations. The other half of the activation route.
3. CYP27A1 (human, Q02318) — secondary 25-hydroxylase / sterol
   27-hydroxylase; pleiotropic (also acts on bile-acid precursors).
4. CYP3A4 (human, P08684) — drug-metabolizing CYP; vitamin D
   25-hydroxylation is a minor, non-core activity. Good test case for
   MARK_AS_OVER_ANNOTATED vs. KEEP_AS_NON_CORE.
5. CYP24A1 (human, Q07973) — 24-hydroxylase / vitamin D inactivator. The
   edge case: decide whether GO:1901755 is the right successor or whether
   a catabolic vitamin D term should be used instead. Coordinate with the
   upstream ontology ticket.

The rodent orthologs (Cyp27b1 mouse/rat, Cyp2r1 mouse, Cyp24a1 rat,
Cyp27a1 rat, cyp27b1 zebrafish) are not initial-priority for individual
review here — the human reviews will pull their orthologs in via IBA, and
curators in MGI/RGD/ZFIN have already marked or can mirror the upstream
remapping.

Proposed approach

1. Wait for the upstream replaced_by to land. The first two ontology
   proposals (xref + has_primary_output) are marked done; the obsoletion
2. replaced_by step is still pending in go-ontology#32077. Once it
   merges, all 24 annotations migrate automatically.
3. Open a CYP27B1 review first (just fetch-gene human CYP27B1) as the
   highest-value, lowest-risk case. Use it to validate that the re-defined
   GO:1901755 is the right molecular_function / involved_in BP for the
   1α-hydroxylation step.
4. Add CYP2R1 and CYP27A1 next; these establish the 25-hydroxylase
   half of the pathway and will also need a vitamin D metabolic / sterol
   hydroxylase MF.
5. Handle CYP3A4 and CYP24A1 as edge cases. For CYP3A4, the vitamin D
   25-hydroxylation is a non-core, low-affinity activity and may warrant
   KEEP_AS_NON_CORE or MARK_AS_OVER_ANNOTATED rather than ACCEPT. For
   CYP24A1, the right successor is plausibly a catabolic vitamin D term,
   not GO:1901755; flag this back to the upstream tracker if confirmed.
6. No InterPro2GO / UniRule / UniProt-Keyword cleanup needed. The
   upstream issue explicitly notes "No Mappings" for GO:0036378. The
   downstream-mapping triage that the larger obsoletion projects need
   (e.g. [[NLS_BINDING_OBSOLETION]], [[VESICLE_DOCKING_OBSOLETION]]) does not
   apply here.

Priority

Low–Medium. Only 24 curated annotations on a five-gene set with no current
reviews here, no mapping fallout, and both upstream curator groups already
marking their work done. The value of opening this is twofold: (a) a clean
test of the replaced_by flow on a small, well-defined pathway; and (b) a
natural anchor for a focused vitamin D / cytochrome P450 mini-review of
CYP27B1 (medically important; vitamin D-dependent rickets type 1A) and its
hydroxylase partners. Skip if higher-priority obsoletions are still open.

Status

• 2026-06-20 — Project file created. Tracking go-annotation#6449 (opened
  2026-06-08, both BHF-UCL and UniProt marked done) and the open
  go-ontology#32077 (third proposal — replaced_by GO:1901755 — still
  pending). The 24 affected curated annotations were retrieved from QuickGO
  (goId=GO:0036378, experimental evidence) and reconciled against the
  upstream group tally. Both GO:0036378 and GO:1901755 confirmed live in
  OLS. No existing AI Gene Review files reference GO:0036378; none of the
  five affected CYP genes (CYP27B1, CYP2R1, CYP27A1, CYP3A4, CYP24A1) has a
  review here yet. CYP24A1 flagged as the one annotation whose successor
  term needs confirmation (catabolic vs biosynthetic vitamin D).
• 2026-09-26 — Repo check: human CYP27A1 now has a review
  (genes/human/CYP27A1/CYP27A1-ai-review.yaml, added in #2003). Its GOA file
  has the PMID:15465040 IDA row as GO:1901755 (no GO:0036378 row remains),
  reviewed as KEEP_AS_NON_CORE. CYP27B1, CYP2R1, CYP3A4 and CYP24A1 still
  have no review.

Slides

• Slides (Marp source: CALCITRIOL_BIOSYNTHESIS_OBSOLETION-slides.md) — AI generated