Estrogen signalling coverage census

Where estrogen signalling is curated in kb/disorders/, counted in six tiers from loosest to strictest. Regenerate with just estrogen-census; the counts move with every curation PR, so treat the numbers here as a dated snapshot and the script as the deliverable. Background: issue #12925.

Entries scanned: 3244.

Tier	Test	Entries	Share
`MENTIONS_ESTROGEN`	names estrogen/estradiol/estrone/estriol or ER-alpha/ER-beta anywhere	118	3.6%
`NODE_INVOKES`	the same, restricted to `pathophysiology:`	58	1.8%
`GO_BOUND`	binds `GO:0030520` estrogen receptor signaling pathway	13	0.4%
`RECEPTOR_NAMED`	names `ESR1`/`ESR2` or the phrase “estrogen receptor”	32	1.0%
`GENE_BOUND`	binds `hgnc:3467` (ESR1) or `hgnc:3468` (ESR2) anywhere	11	0.3%
`GENE_ON_NODE`	binds either gene in a `genes:` descriptor on a pathophysiology node	8	0.2%

The tiers are not strictly nested. An entry can bind the pathway without naming the receptor, and can name the receptor without binding the pathway, so each row is counted independently.

Drift since commit `94591e52d4`

The baseline column is the figure reported in issue #12925 when the gap was described. Any difference is ordinary curation drift, except as noted below the table.

Tier	Baseline	Now	Change
`MENTIONS_ESTROGEN`	115	118	+3
`NODE_INVOKES`	57	58	+1
`GO_BOUND`	12	13	+1
`RECEPTOR_NAMED`	29	32	+3
`GENE_BOUND`	6	11	+5
`GENE_ON_NODE`	1	8	+7

One of those differences is not drift. RECEPTOR_NAMED was measured with a line-based search, which misses CHEK2-related_Cancer_Predisposition: it writes “oestrogen receptor” in a folded scalar, so the phrase is split across two source lines and only exists once the document is parsed. Re-measuring the baseline tree with this script gives 30, not 29.

The gap this was written to measure

6 entries bind GO:0030520 on a node while putting no receptor gene on any node. The pathway is annotated and the receptor driving it is not.

Entry	Node(s) binding the pathway
`Aromatase_Deficiency`	Estrogen Deficiency and Androgen Excess
`Breast_Carcinoma`	ER and HER2 Receptor Heterogeneity
`Breast_Fibroadenoma`	Dysregulated Stromal Estrogen Signaling and ECM Organization
`PIK3CA_Mutant_Breast_Cancer`	Endocrine Resistance
`Testicular_Sex_Cord_Stromal_Neoplasm`	Estrogen-Mediated Clinical Manifestations
`Triple_Negative_Breast_Cancer`	Loss of Hormone Receptor Signaling

Each row is a research task, not a mechanical backfill. A genes: descriptor asserts that this receptor drives this mechanism in this disease, and some of these nodes are loss-of-signalling or ligand-supply claims where an ESR1 binding would be wrong. Deciding not to bind, with the reason recorded in notes, is a closed task.

Receptor bound, but off the pathograph

3 entries bind ESR1 or ESR2 somewhere without the gene reaching a pathophysiology node. These are typically susceptibility polymorphisms in genetic:, where a genotype-association paper may legitimately stop short of supporting a causal edge.

Entry	Gene(s) bound
`Breast_Carcinoma`	ESR1
`Premenstrual_Dysphoric_Disorder`	ESR1
`Skeletal_Fluorosis`	ESR1

Receptor on the pathograph

46_XX_Gonadal_Dysgenesis — ESR2
ER_Positive_Breast_Cancer — ESR1
Endometrial_Carcinoma — ESR1
Endometrial_Endometrioid_Adenocarcinoma — ESR1
Estrogen_Resistance_Syndrome — ESR1
Heart_Failure — ESR1, ESR2
Lymphangioleiomyomatosis — ESR1
Osteoporosis — ESR1

Modules

Modules that mention estrogen or the receptor, and whether they bind it. “In module prose” separates a module that models the receptor from one that merely cites a paper about it: a mention confined to reference_title or snippet is quoted source metadata, not a claim the module makes. No module binds either gene. A module node binding is a design decision rather than a backfill, because a generic receptor binding may belong only in the conforming entries.

Module	In module prose	Binds `GO:0030520`	Binds ESR1/ESR2
`cdk46_inhibitor_resistance`	yes	no	no
`cholestatic_liver_injury`	yes	no	no
`deep_placentation_defect`	quoted sources only	no	no
`deregulated_nutrient_sensing`	yes	no	no
`osteoporosis_bone_resorption`	yes	no	no
`sex_steroid_driven_proliferation`	yes	yes	no