Genomics researcher defends AI variant prioritization paper: validation already sufficient

anshulkundaje · x · 2026-09-18

In an ongoing thread about a disease variant prioritization paper, genomics researcher Anshul Kundaje acknowledges that more experimental validation and direct testing is needed — but argues the paper's presented evaluation is more than sufficient to support its claims: that the score is likely a useful variant prioritization tool and the resource serves as a useful lookup table for anyone.

The exchange reflects a broader debate over how much experimental validation computational scoring tools require before publication.

Related event: Stanford expert defends AlphaGenome variant scores in detailed thread, rebutting "AI slop" criticism(20 posts)→

Original post →

More from Research

Research channel →