- | 4:00 pm
How Google DeepMind is helping researchers to more efficiently catalog genetic mutations
Google DeepMind’s AlphaMissense system is helping researchers focus on therapies for major diseases.

In 2023, Google DeepMind adapted its landmark AlphaFold protein-modeling AI to create a new system called AlphaMissense, which classifies “missense variants” (genetic mutations that can affect the function of proteins in our cells) as either benign or potentially dangerous.
Traditional laboratory methods have been able to classify only a tiny fraction of known missense variants; the work is expensive and time-consuming, because each affected protein is unique and must be studied using distinct experiment designs. Trained on a variety of known human and primate missense variants, AlphaMissense predicted and categorized 89% of all 71 million possible missense variants. Its predictions were shown to be state-of-the-art after classifying variants in a range of common benchmark tests.
With the classifications now cataloged and available, researchers can focus their time and money on missense variants that they’re reasonably sure could be associated with such serious diseases as cystic fibrosis, sickle cell anemia, and cancer.
That research could lead to faster diagnoses and more effective therapies informed by the human genome.
Explore the full 2024 list of Fast Company’s Most Innovative Companies, 606 organizations that are reshaping industries and culture. We’ve selected the firms making the biggest impact across 58 categories, including advertising, artificial intelligence, design, sustainability, and more.