Google has introduced the Google Antigravity platform, an intelligent development environment designed to accelerate the discovery of new antibiotics. The system integrates scientific workflows into a single session, allowing biologists to search, evaluate, and screen potential targets without switching between multiple databases.
The platform utilizes the ProtNLM deep learning model to analyze millions of uncharacterized protein sequences from the UniProt database. By interpreting the evolutionary grammar of amino acid sequences, the model predicts biological functions directly, helping to identify promising candidates for antimicrobial peptides.
Researchers are currently using Antigravity to develop a new antimicrobial peptide candidate aimed at respiratory pathogens, which is now undergoing wet lab validation. The platform enables collaboration between scientists and AI agents, where the agents manage data coordination and literature synthesis while researchers define biological goals.
Antigravity includes a 3D visualization tool called Biomolecular Structure Viewer. This tool allows users to pull structural data from the AlphaFold database, highlight functional regions, and interact with the AI to analyze molecular structures in a shared 3D space, replacing the need for fragmented software tools.
