AI for Biosecurity and Drug Discovery

8 posts · May 2026 – Sep 2026

What this trail covers

AlphaGenome Atlas is introduced, a platform containing predictions for the effects of 9 billion single-nucleotide variants in the human genome. It includes the AlphaGenome Variant Impact (AVI) score, which combines AlphaGenome and AlphaMissense predictions, and AVI feature attributions to explain the drivers of the score. The platform is available via a website portal, API, and as a skill in Google Antigravity. It is a massive 1-petabyte dataset, expanding on previous work with AlphaFold.

All DeepMind trails

Sep 2026AlphaGenome Atlas: Molecular predictions for 9 Billion human DNA variantsThis post introduces AlphaGenome Atlas, a 1-petabyte dataset containing predictions for the molecular impact of 9 billion single-nucleotide variants in the human genome. It details the creation of the AlphaGenome Variant Impact (AVI) score, which combines predictions from AlphaGenome and AlphaMissense models, and introduces AVI feature attributions to explain the drivers of these scores. The post highlights the application of this resource in identifying variants for rare disease research and mapping rare variants associated with complex traits, demonstrating a significant expansion in the scale and accessibility of genomic variant prediction.
May 2026Co-ScientistIntroduced Co-Scientist, an AI tool that assists researchers in hypothesis generation and refinement for disease research. Co-Scientist processes grant proposals and unpublished data to identify promising hypotheses and prioritize specific proteins and amino acids for experimental investigation, significantly accelerating the research timeline from years to months.
May 2026Co-ScientistIntroduced Co-Scientist, an AI system used to accelerate drug discovery by analyzing scientific literature to identify potential drug candidates for repurposing. Demonstrated its effectiveness in identifying drugs for liver fibrosis treatment, with two out of three suggested candidates showing positive results in lab tests, including the cancer drug vorinostat. This highlights Co-Scientist's ability to suggest drugs that reshape gene activity, potentially leading to new anti-fibrotic medicines.
May 2026Co-ScientistIntroduced Co-Scientist, an AI system used to connect scattered findings in the biology literature, generate novel hypotheses, and refine experimental design for testing those hypotheses. Demonstrated its application in aging research by generating a hypothesis about ISR regulation by metabolism and aiding in experimental design.
May 2026Co-ScientistIntroduced Co-Scientist, an AI system used to comb biomedical literature for overlooked links and generate new hypotheses. Demonstrated its application in identifying candidate combination therapies for metabolic dysfunction-associated steatohepatitis (MASH) by synthesizing evidence across liver biology and pharmacology, and highlighting the NLRP3 inflammasome as a key molecular bridge.
Advertisement
May 2026Co-ScientistIntroduced Co-Scientist, an AI system that assists in identifying novel genetic factors for reversing cellular senescence and accelerates the analysis of large-scale genetic screening data by integrating scientific literature, reducing analysis time from months to days.