Daijiworld Media Network - London

London, Jan 31: Google DeepMind has unveiled a new artificial intelligence tool, AlphaGenome, aimed at helping scientists identify genetic changes that trigger diseases and accelerate the development of new treatments.

According to DeepMind, AlphaGenome can analyse up to one million DNA letters at a time and predict how mutations influence gene regulation — including when genes are switched on, which cells they act in, and how strongly they function.

Experts say this is crucial because many common inherited conditions such as heart disease, autoimmune disorders, mental health illnesses and several cancers are often linked not to protein-making genes, but to regulatory changes that control gene behaviour — an area that has long remained difficult to decode.

DeepMind researcher Natasha Latysheva said AlphaGenome is designed to improve understanding of how the genome operates and to speed up research into the biological code that governs life.

The human genome carries nearly three billion DNA letter pairs, but only about two percent directly codes for proteins. The remaining DNA largely controls where, when and how much genes are activated, shaping how the body functions and responds to disease.

AlphaGenome has been trained using publicly available human and mouse genetic datasets, enabling it to connect specific mutations with changes in gene activity across different tissues.
The tool can also predict mutation-driven effects in organs such as the brain and liver, researchers said.

Scientists believe AlphaGenome could play a key role in identifying genetic regions involved in cancer and other complex diseases, and may assist in designing gene therapies by helping researchers create DNA sequences that activate genes in specific cell types.

Independent experts have described AlphaGenome as a significant breakthrough, while cautioning that more research is needed before AI-based predictions can reliably replace laboratory testing.