AI scores 27M genomic variants across 1,500 cell states, ~40B effects via ENCODE GRAMMAR

anshulkundaje · x · 2026-09-15

At the IGVF meeting at Washington University in St. Louis, changmyun's team presented work using ENCODE GRAMMAR to score 27M candidate variants across 1,500 cell states — roughly 40 billion variant-effect predictions — with 4.5M variants prioritized so far. A large-scale example of AI applied to genomics.

Original post →

More from Research

Research channel →