Sign in Subscribe
Peptides

Breakthrough in Antimicrobial Peptide Discovery against Escherichia coli

Strategic Insights

— 1 min read

Researchers at the University of the Philippines have developed a novel machine learning model, ISCAPE, to accelerate the discovery of antimicrobial peptides (AMPs) against Escherichia coli. ISCAPE outperforms existing predictors, achieving an impressive area under the receiver operating characteristic curve (AUROC) of 91.83% and a Matthew's correlation coefficient (MCC) of 71.86%. The model's interpretability is enhanced through SHapley Additive exPlanations (SHAP), which identifies the molecular features most critical for AMP activity.

Key Takeaways:

  • ISCAPE is a machine learning model developed to predict the activity of both natural and chemically modified peptides against E. coli ATCC 25922.
  • The model requires only a Simplified Molecular-Input Line-Entry System (SMILES) string as input and can predict the activity of peptides against E. coli with a minimum inhibitory concentration (MIC) threshold of 16 mg/mL.
  • ISCAPE outperformed the state-of-the-art AntiMPmod, achieving an AUROC of 91.83% and a MCC of 71.86%.
  • The model's performance is driven by features including the fraction of carbon-carbon pairs and extended connectivity fingerprints (ECFPs).
  • ISCAPE's interpretability is enhanced through SHAP, which identifies the molecular features most critical for AMP activity.
  • The research has been peer-reviewed and published in the Journal of Molecular Graphics and Modelling.
  • The model is a user-friendly tool that allows experimentalists to pinpoint key molecular features, reducing the need for extensive structure-activity relationship (SAR) studies and guiding the design of novel AMPs.

Statistics:

  • AUROC of 91.83% achieved by ISCAPE in predicting AMP activity against E. coli.
  • MCC of 71.86% achieved by ISCAPE in predicting AMP activity against E. coli.
  • 16 mg/mL as the minimum inhibitory concentration (MIC) threshold for predicting AMP activity against E. coli.
  • 142:109188 as the journal reference for the peer-reviewed article published in the Journal of Molecular Graphics and Modelling.
  • 2025 as the year the research was published.
  • 1 as the reference number for the Journal of Molecular Graphics and Modelling in the list of sources.

Sources:

  • Journal of Molecular Graphics and Modelling

Journal of Molecular Graphics and Modelling, 2025;142:109188.

  • Remmer L. Salas

Institute of Chemistry, College of Science, University of the Philippines, Diliman, Quezon City, 1101, Philippines.

  • Portia Mahal G. Sabido

Institute of Chemistry, College of Science, University of the Philippines, Diliman, Quezon City, 1101, Philippines.

  • Ricky B. Nellas

Institute of Chemistry, College of Science, University of the Philippines, Diliman, Quezon City, 1101, Philippines.

Read more