#1
#1 out of 1
technology1d ago
How DeepMind's genome AI could help solve rare disease mysteries
- AlphaGenome was used at an Undiagnosed Hackathon to tackle 29 conditions, aiming to expand diagnostic options beyond coding regions.
- The hackathon, organized by the Wilhelm Foundation, brought researchers to the Mayo Clinic and Europe to diagnose undiagnosed conditions.
- AlphaGenome predicts non-coding mutations' effects on gene expression, aiding interpretation beyond the exome.
- Early tests linked a non-coding variant to a patient’s diagnosis, with lab work supporting the AI’s predictions.
- The effort highlights a new approach to diagnosing rare diseases by exploring the genome’s non-coding regions.
- The hackathon participants aim to help the roughly 350 million people with undiagnosed rare conditions.
- The Wilhem Foundation supports families affected by undiagnosed diseases and organized the event.
- The study situates AlphaGenome within efforts to interpret non-coding variants as a complement to exome testing.
- The Nature report notes AlphaGenome’s ability to predict non-coding impacts could broaden diagnostic avenues.
- Researchers highlight the need for integrating AI with laboratory validation to confirm predictions.
Vote 0