Observed arrival · 2026-10-13
Manuel Corpas’s case for trustworthy agentic genomics
A genomics scientist’s personal site traces his work on genomic equity and presents open tools for evaluating AI agents that perform genomic analysis.
- For
- Genomics researchers and builders of AI analysis agents
- Worth noticing
- The page identifies ClawBench’s reference basis as Genome in a Bottle and says its scoring rewards abstention and provenance.
Field notes
The page frames genomic interpretation as a problem of both evidence and representation: it says 86% of genome-wide association data still comes from people of European ancestry. Its project links separate agent skills from evaluation: ClawBio provides the described analysis layer, while ClawBench and GenomeQA assess answers, abstention, provenance and population robustness. A linked demonstration is described as using a real 30x whole genome.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue