Researchers at the University of California San Diego have used AI to decode a fundamental control element in the genome: the “initiator” – a kind of hidden on switch that determines when and where genes become active. It is found in about 60 percent of human genes.
The finding
The team led by Professor James T. Kadonaga, with graduate researcher Torrey Rhyne-Carrigg, generated and analyzed around 500,000 variant DNA sequences using high-throughput sequencing. With this data, the researchers trained a machine-learning model that reliably recognizes the base-pair patterns of functional initiators.
Why it matters
The model identified the initiator in about 60 percent of human genes – a long-standing puzzle in molecular biology. The breakthrough could help predict the effects of harmful mutations and enable the targeted design of synthetic promoters: to switch genes precisely on and off for biotechnology and therapy.
Sources: ScienceDaily, phys.org.
