Stats
Coverage, discovery confidence, audit-flag distribution, and compute spend for the genome-wide run.
19,291
Genes annotated
99.97% of 19,296
274,637
Discoveries
avg 14.2 per gene
6,049
Round 2 retried
31.4% of genome
$4,934
Total spend
$0.256/gene
Discovery confidence
Tier × preponderance · 274,637 findings
Assigned per-finding by the reading pass from method tier and weight of evidence. High = direct biochemical/structural/genetic proof; Low = single-paper observational claim.
Audit flags
Narratives the deterministic concordance detector held for revision
| Tier 1 — wrong gene | 17 |
| Tier 2 — rerun needed | 134 |
| Tier 3 — alt product | 48 |
| Total flagged | 199 (1.0%) |
Six rules (R4–R9) collapse into the three tiers above by what action each implies: wrong gene needs a rewrite, rerun needed is a refusal that can be re-prompted, alt product describes a non-coding RNA the protein-pipeline shouldn't have annotated.
Discovery year histogram
274,339 dated findings · 1980 → 2026 · post-2020 in red
1980
2003
2026
≥ 2020: 79,721 findings (29.1% of dated).
Compute ledger
USD · batch rates · April 2026 run + Round 2
Round 1 — full genome
$3,886.71
19,291 genes · ~$0.20/gene
Round 2 — R1∪R2∪R3 retry
$1,047.27
6,049 flagged genes
Total LLM spend
$4,933.98