Molecular de-extinction of ancient antimicrobial peptides enabled by machine learning

Abstract

Molecular de-extinction could offer new avenues for drug discovery by reintroducing bioactive molecules that are no longer encoded by extant organisms. To prospect for antimicrobial peptides encrypted as subsequences of extinct and extant human proteins, we introduce the panCleave random forest model for proteome-wide cleavage site prediction. Our model outperformed multiple protease-specific cleavage site classifiers for three modern human caspases, despite its pan-protease design. Antimicrobial activity was observed in vitro for modern and archaic protein fragments identified with panCleave. Lead peptides were tested for mechanism of action, resistance to proteolysis, and anti-infective efficacy in two pre-clinical mouse models. These results suggest that machine learning-based encrypted peptide prospection can identify stable, nontoxic antimicrobial peptides. Moreover, we establish molecular de-extinction through paleoproteome mining as a framework for antibacterial drug discovery. HighlightsO_LIMachine learning guides bioinspired prospection for encrypted antimicrobial peptides. C_LIO_LIModern and extinct human proteins harbor antimicrobial subsequences. C_LIO_LIAncient encrypted peptides display in vitro and in vivo activity with low host toxicity. C_LIO_LIPaleoproteome mining offers a new framework for antibiotic discovery. C_LI