Jan 30, 2026
3 min read
AlphaGenome can analyze up to 1m letters of DNA code simultaneously and can pave the way for new treatments.
Researchers at Google DeepMind have unveiled their latest artificial intelligence tool that could help scientists identify the causes of disease and develop new treatments.
AlphaGenome predicts how mutations interfere with gene regulatory pathways, when mutations are activated, where cells in the body are, and how much their biological control is up or down.
The most common diseases running in families, including heart disease and autoimmune disorders, as well as mental health problems, like many cancers, are associated with mutations that affect genes regulation, but it is not always easy to say what genetic defects are to blame.
“We view AlphaGenome as a tool for understanding what functional elements of the genome do, which we hope will accelerate our fundamental understanding of the code of life,” DeepMind researcher Natasha Latysheva said at a press conference about her work.
There are 3 billion pairs of letters in the human genome - including GS, DS, CS and the DNA code.2% of the genome tells cells how to make proteins, the building blocks of life.The rest organizes gene activity, which contains important instructions that tell where, when and how much individual genes change.
The researchers trained AlphaGenome on public databases of human and mouse genetics, allowing it to learn relationships between mutations in specific tissues and their effect on gene regulation.The artificial intelligence can analyze up to 1 million letters of DNA code simultaneously and predict how mutations will affect various biological processes.
The DeepMind team believes this tool will help researchers uncover which genes are most important for the development of certain tissues, such as nerve and liver cells, and identify the key changes that drive cancer and other diseases.It could also support new genetic therapies by allowing researchers to design entirely new DNA sequences – for example, changing a gene in nerve cells but not in muscle cells.
Carl de Boer, a researcher at the University of British Columbia in Canada who was not involved in the work, said: "AlphaGenome can determine if changes affect the regulation of the genome, which genes are affected and how, and which types of cells.
"Ultimately, our goal is to have models that are so good that we don't need to do an experiment to confirm their predictions. Although AlphaGenome represents a significant innovation, it will require continued work from the scientific community to achieve this goal."
Some researchers have started using AlphaGenome.Marc Mansour, associate professor of pediatric hematos-oncology at UCL, said it marked a "turning point" in his work to find genetic drivers for cancer.
Gareth Hawkes, a statistical geneticist at the University of Exeter, said: "The non-coding genome makes up 98% of our 3 billion base pair genome. We understand this 2% very well, but the fact that we have AlphaGenome that can predict this other 2.94 billion base pair region is a big step for us."
