DBAASP V4: condition-resolved bioactivity and chemical structures for data-driven peptide and peptidomimetic discovery.

DBAASP v4 expands the structure-aware Database of Antimicrobial Activity and Structure of Peptides to over 25 000 entries, adding SMILES, antibiofilm/synergy records, and >9200 molecular-dynamics trajectories.

• Why it matters: Machine learning for peptide and peptidomimetic discovery needs experimentally anchored, condition-resolved labels, but large candidate catalogs lack the quantitative context required for supervised modeling.
• What they did: The team manually curates peptides and peptidomimetics with quantitative antimicrobial, antibiofilm, anticancer, hemolytic, and mammalian-cell cytotoxicity measurements plus experimental conditions, and adds structure annotations and PubChem links.
• The result: The condition-resolved, structure-rich dataset enables strain- and target-specific supervised learning, calibration, and surrogate fitness functions for genetic algorithms, supporting testable prediction and optimization.

AI summary of the abstract · may contain errors · not medical advice · read the paper.