GoKawiilQuanta Magazine: AI gene-regulation models like Evo 2, Genos and AlphaGenome still fall short of explaining the genome
A Quanta Magazine report reviews decades of genome research since the 1950s discovery of DNA's structure, noting that the Human Genome Project found only about 2% of human DNA encodes genes. The bulk of the mystery now lies in gene regulation—how genes are switched on and off—and researchers are turning to AI 'foundation models' such as Evo 2, Genos and Google DeepMind's AlphaGenome to help predict how DNA sequence changes affect biology.
GoKawiil's interpretation of the reporting above, not reported fact.
The report suggests that gene regulation may be too complex for traditional human analysis alone, which could explain why biologists are increasingly relying on AI tools trained on massive genomic datasets. This points to a broader shift in biology toward AI-assisted modeling, though the piece implies such models are not yet capable of fully capturing the genome's regulatory complexity.
- Only about 2% of the human genome consists of protein-coding genes, per the Human Genome Project.
- Gene regulation—not gene identification—is now seen as biology's central unsolved puzzle.
- AI foundation models like Evo 2, Genos and AlphaGenome are being tested to predict effects of DNA sequence variation.
Source: wired.com — Philip Ball, 2026-10-17
Published there as: “Why AI Isn’t Ready to Tackle Our Convoluted Genomes”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.