Article body
Researchers at the University of California, San Diego have used AI to uncover the DNA signature of a key genetic "switch" known as the initiator, which is involved in turning genes on. By analyzing approximately 500,000 DNA sequences, the model identified the initiator in about 60% of human genes. This breakthrough could help predict the effects of harmful mutations and contribute to understanding the broader genetic instructions that control gene activity throughout the body.
The study, led by graduate student researcher Torrey Rhyne-Carrigg, utilized high-throughput DNA sequencing to measure gene expression activity across various versions of the initiator. The researchers then trained a machine learning system to identify the characteristic DNA pattern associated with the initiator. Once decoded, the model was able to predict the presence or absence of the initiator in human genes for the first time.
The findings may help researchers anticipate how mutations affecting the initiator could alter gene activity and contribute to various disorders. Additionally, the study's data and AI models could support the design of synthetic promoters that can switch genes on or off for specific functions. The research highlights the potential of combining laboratory experiments with AI to uncover information encoded in human DNA.