Latest / Health and life sciences

Google DeepMind’s AlphaMissense classifies 71 million possible disease-linked mutations

See how this story connects

Google DeepMind published AlphaMissense in Science on 19 September 2023 with a catalogue of predictions for 71 million possible missense variants in human proteins. The model classed 89% of them as likely benign (57%) or likely pathogenic (32%), against 0.1% confirmed by human experts. DeepMind released the predictions free of charge and the model code on GitHub, and moved the predictions to a CC BY 4.0 licence in March 2024.

Why it matters

Clinical geneticists and rare-disease researchers gained free predictions for variants that had no expert classification, which can help narrow down the cause of a patient’s disease.

When
19 Sep 2023
ID
WTA-N-2023-0093
Who
Google DeepMind

Summary by wheresthe.ai, from Google DeepMind and the reporting credited above. Checked against the source on 9 Oct 2026.

Source copy: download the page as saved on 18 Oct 2026 (PDF). Also held by the Internet Archive.

Earlier in this Storyline: What led hereMoorfields and UCL publish RETFound, an open foundation model trained on 1.6m retinal images

Cite this story

WTA-N-2023-0093: “Google DeepMind’s AlphaMissense classifies 71 million possible disease-linked mutations.” wheresthe.ai, 19 Sep 2023. https://wheresthe.ai/d/deepmind-alphamissense-2023-09/ (source: Google DeepMind).

Spotted an error? Email [email protected].