Google releases AlphaGenome Atlas predicting effects of all 9 billion single-base DNA variants

September 9, 2026 at 09:34 PM2 min

This digest was compiled by AI from multiple sources — links to the originals are below.

Google announced AlphaGenome Atlas on Tuesday, a resource predicting the consequences of every possible single-base variant in the human genome. The system evaluates 9 billion DNA bases through AlphaGenome software, focusing on non-coding DNA that makes up over 97 percent of the genome. The release comes as biologists continue to struggle with identifying functional elements in non-coding regions.

Key Facts

AlphaGenome Atlas evaluates 9 billion DNA bases — the three alternative bases at each of the 3 billion positions in the human reference genome.
The system focuses on non-coding DNA, which constitutes more than 97 percent of the human genome.
Google announced AlphaGenome Atlas on Tuesday, according to Ars Technica.
AlphaGenome is designed to identify potential functions of non-coding DNA, including regulatory sequences that control gene activity.

AlphaGenome Atlas Release

Google announced AlphaGenome Atlas on Tuesday, a resource that attempts to predict the consequences of every possible single-base variant in the human genome. The human genome is about 3 billion bases long, so trying the other three DNA bases that don't appear in the reference genome means sending a total of 9 billion bases through AlphaGenome software. AlphaGenome is designed to identify potential functions of non-coding DNA, which does not encode proteins but makes up the vast majority of the human genome. Some non-coding DNA is essential for controlling the activity of the protein-coding portion, telling the cell where and when to make messenger RNAs and how to process them into mature protein-coding forms.

Non-Coding DNA Challenge

The portion of the human genome that encodes proteins is less than 3 percent, while most of the genome is non-coding and contains centromeres, caps that protect chromosome ends, regulatory DNA, and signals for messenger RNA processing. Most non-coding DNA is junk — the remains of ancient viral infections, DNA-level parasites, and genes inactivated by mutation. Proteins that interact with DNA are not very selective about sequences, potentially binding at random throughout the genome and tolerating a certain degree of mutation. Many DNA-binding proteins are cell-type specific, with different populations in liver cells, nerve cells, and immune cells. Google developed the AlphaGenome AI system because evaluating non-coding DNA function involves imprecise probabilities that rely heavily on context, a problem AI is well-suited to solve.