Google DeepMind Uses Antigravity Agent Platform to Find New Antimicrobial Peptide Candidate
In short: Google DeepMind demonstrated how its agentic platform Google Antigravity, combined with the ProtNLM model, helped researchers search millions of uncharacterized proteins in UniProt to find a novel antimicrobial peptide candidate against respiratory pathogens. The AI agent handled database queries, structural comparison via AlphaFold, and genomic context analysis, while a human researcher guided the scientific strategy. The resulting candidate is now undergoing wet-lab validation, and DeepMind introduced a new 3D Biomolecular Structure Viewer built into Antigravity for shared human-agent inspection of protein structures.
This summary was generated automatically by AI from Google DeepMind's publication. It is our own text, not a copy of the original — facts, figures and quotes belong to the source, linked above and below.
What changed?
- 1Agent scans millions of UniProt protein sequences using ProtNLM to predict function from raw amino acid sequences
- 2Agent cross-references candidates against Swiss-Prot/BLAST for novelty and AlphaFold for structural similarity
- 3Agent reasons over genomic neighborhood (operon structure) to support functional hypotheses
- 4Agent generates literature-based mechanism hypotheses and a wet-lab experimental protocol
- 5New Biomolecular Structure Viewer added to Antigravity for synchronized human-agent 3D structure annotation
- 6A novel bacteriocin candidate (UniProt ID A0A0U0D591) identified and now in wet-lab testing
Why it matters
It illustrates how multi-turn agentic workflows with tool use (database queries, structural modeling, literature synthesis) can compress a traditionally fragmented, multi-tool research process into a single collaborative session, a pattern relevant to any team building complex tool-calling agents.
What it means for AI agents and contact centers
This is a domain-specific scientific research tool with no direct voice, telephony, or contact-center application, but the underlying pattern—an agent orchestrating multiple external tools and data sources during a multi-turn session while a human steers strategy—mirrors the architecture used in building tool-calling AI agents more broadly.
Sources
- Google DeepMindOfficialPrimary sourceOriginal article →„Accelerating the search for antimicrobial peptides with Google Antigravity“15 Oct 2026, 19:08
- Published by source
- 15 Oct 2026, 19:08
- Found by our system
- 15 Oct 2026, 19:11
- Summary generated
- 15 Oct 2026, 19:15
This article was written by AI from the original source. Facts, numbers and prices come from the source; missing values are marked “Not specified”. Legal notice, copyright and privacy