ESR-ECOLI-DET-Mini Recap Results
Archived result bundle from PR #1535:
- PR: https://github.com/ai4curation/ai-gene-review/pull/1535
- PR title:
DO NOT MERGE Independent reviews of 7 E. coli genes + DeepECTF prediction assessments (claude-expt-1) - Source head commit:
9ce787b0daddbf03d6bc1b5164682f7e0000e7cd - Dataset ID:
10.5281/zenodo.20751016 - Benchmark ID:
ESR-ECOLI-DET-Mini(alias:ESR-ECOLI-DET-7)
These files are preserved here as project experiment outputs only. They should not be treated as production updates to genes/ECOLI/.
Contents
genes/ECOLI/DET-PREDICTION-REVIEW-SUMMARY.md— original experiment summary.genes/ECOLI/<gene>/<gene>-det-predictions-review.yaml— DeepECTF prediction review outputs.genes/ECOLI/<gene>/<gene>-ai-review.yaml— independently recreated gene reviews from the experiment.genes/ECOLI/yciO/yciO-bioinformatics/— YciO/TsaC alignment artifacts used by the experiment.publications/PMID_*.md— supporting publication caches added by the experiment branch.
ESR-ECOLI-DET-Mini Verdicts
| Gene | DeepECTF prediction | Experiment verdict | Error type |
|---|---|---|---|
| fepE | EC:2.7.13.3 histidine kinase | REP | FREQUENCY_BIAS |
| yciO | EC:2.7.7.87 TC-AMP synthase | PLI | PARALOG_OVERANNOTATION |
| yegV | EC:2.7.1.92 5-dehydro-2-deoxygluconokinase | UNC | |
| ygfF | EC:1.1.1.47 glucose 1-dehydrogenase | UNC | |
| yjdM | EC:3.11.1.2 phosphonoacetate hydrolase | NPI | IN_VITRO_NOT_IN_VIVO |
| yjhQ | EC:2.3.1.189 mycothiol synthase | NPI | PATHWAY_CONTEXT_IGNORED |
| yrhB | EC:4.1.2.50 6-carboxytetrahydropterin synthase | NPI | PATHWAY_CONTEXT_IGNORED |
Relative to the published de Crécy-Lagard expert labels, this experiment is an exact class match for fepE, yciO, yjhQ, and yrhB, and differs for yegV, ygfF, and yjdM. It is therefore a 4/7 exact-label Expert Synthetic Review recap, not a 7/7 replication.
Note on wrong reference identifiers (2026-09-27)
The production DeepECTF reviews (genes/ECOLI/*/*-det-predictions-review.yaml) cited five wrong-identifier PMIDs. They were corrected on 2026-09-27 in commit 07b082cd2:
| Cited as | What the PMID actually resolves to | Correction in production files |
|---|---|---|
| PMID:37820725, the DeepECTF paper | a labile-iron imaging study | replaced by PMID:37963869 (Kim et al. 2023) |
| PMID:9325100 | a serum amyloid P paper | replaced by PMID:9300819 (Kulakova et al. 1997) |
| PMID:21135036 ("Gerdes 2011") | a dental age table | removed; the intended paper could not be identified |
| PMID:28194017 ("Zallot 2017 QueD") | a bismuth melting study | removed; the intended paper could not be identified |
| PMID:8335646 ("Metcalf & Wanner 1993") | a menaquinone paper | removed; the intended paper could not be identified |
These blinded experiment outputs are archived as generated and were not edited. A search of this folder on 2026-09-27 found none of the five wrong identifiers. The copies here already cite PMID:37963869 for DeepECTF, so no correction applies to them. This note records the check so that the recap and production files are not confused.