DeepMind releases a genome map covering all human gene mutations

Google DeepMind released a new genome “map” on September 8, 2026. Using the AlphaGenome AI model, it predicted the effects of approximately 9 billion single nucleotide mutations in the human genome and included over 100 million short insertion or deletion variations. This map aims to assist in the diagnosis of rare diseases, reveal hidden mechanisms of common diseases, and understand DNA regulatory rules. The total amount of data involved is 1 petabyte. The AlphaGenome Variant Impact (AVI) scoring metric developed by the team is used to quantify the biological impact of variants. Currently, this tool is available for non-commercial use free of charge, and it has attracted approximately 9,000 researchers who have used it through the API. However, some biologists face difficulties in using it due to the need to write code.