Google's AI genome system evaluates every possible one-base change
Summary
Google created AlphaGenome Atlas, a tool using artificial intelligence to predict how every possible one-letter change in the human DNA might affect gene function. This system focuses on the non-coding parts of DNA, which do not make proteins but help control gene activity. The tool can help researchers understand which DNA changes might be important.Key Facts
- The human genome has about 3 billion DNA bases. AlphaGenome checks all possible changes by testing three alternative bases at each spot, totaling 9 billion changes.
- Most of the human genome is non-coding DNA, meaning it does not produce proteins but controls when and where genes are active.
- Non-coding DNA includes parts like centromeres (which help chromosome division), protective caps, and regulatory sequences that manage gene expression.
- Much non-coding DNA is “junk” from ancient viruses or inactive genes. Scientists find it hard to tell which parts are functional.
- AlphaGenome uses AI to predict functions like gene expression, protein binding sites, and how DNA is packaged in cells.
- The system is trained using data from humans and mice and focuses on cell types that have been well studied.
- It helps researchers decide whether DNA changes in non-coding areas might be meaningful and suggests reasons why.
- AlphaGenome’s predictions perform as well or better than other specialized tools used by scientists.
Read the Full Article
This is a fact-based summary from The Actual News. Click below to read the complete story directly from the original source.
