AlphaGenome 2: How Google DeepMind Taught an AI to Read DNA Through a Cell's Eyes
For decades, one of the hardest problems in human genetics has been connecting a DNA variant to the specific cell type where it actually does something. Genome-wide association studies (GWAS) can flag thousands of variants linked to disease, but most of them sit in non-coding regions β the 98% of the genome that doesn't encode proteins β and their effects depend entirely on which cell is reading that stretch of DNA. A variant that disrupts gene regulation in a gut immune cell may be invisible in a liver cell.
Google DeepMind's AlphaGenome 2 , announced on November 11, 2026, is a direct attempt to solve this problem. It's a conditional DNA sequence model that predicts how genetic variants affect molecular activity in specific human and mouse cell types β not just broad tissue averages.β¦