alphamissense

is a Data Source.

AlphaMissense is a deep learning model from Google DeepMind that predicts the pathogenicity of missense variants across the human proteome. It provides precomputed pathogenicity scores and classifications (likely benign, likely pathogenic, or ambiguous) for all approximately 71 million possible single amino acid substitutions in the human genome, as well as substitution effect scores for the entire proteome. The predictions are distributed via bulk download and accessible through hosted viewers and genome browsers.

License

CC BY 4.0

Homepage

alphamissense

Repository

GitHub

Infores ID

Unknown

FAIRsharing ID

Unknown

Product Summary

Products

From this Resource
ID Name URL Category Format Description
alphamissense.predictions AlphaMissense Variant Predictions 10813168 Product tsv Precomputed AlphaMissense pathogenici...
alphamissense.software AlphaMissense Software alphamissense ProcessProduct python Open-source implementation of the Alp...
alphamissense.viewer AlphaMissense Hosted Viewer alphamissense.hegelab.org GraphicalInterface http Web-based viewer for exploring AlphaM...
alphamissense.docs AlphaMissense Documentation alphamissense DocumentationProduct http Project overview, methodology, and li...
From other Resources
ID Name URL Category Format Relation Description
biobtree.api BioBTree REST API api ProgrammingInterface http had primary source REST API for searching identifiers an...

Details

AlphaMissense

AlphaMissense is a deep learning model developed by Google DeepMind that predicts the pathogenicity of missense variants across the human proteome. Building on the protein structure modeling approach of AlphaFold, it classifies single amino acid substitutions as likely benign, likely pathogenic, or ambiguous, and assigns a continuous pathogenicity score to each variant.

The resource provides precomputed predictions for all approximately 71 million possible human missense variants, distributed as bulk download files via Zenodo and Google Cloud Storage and accessible through hosted viewers and genome browsers. In KG-Registry, the AlphaMissense products point to the variant prediction dataset, the open-source prediction software, a hosted exploration viewer, and project documentation.

Is this information incorrect or incomplete? Request an update.

Created: June 15, 2026 | Last modified: June 15, 2026