UP chemists engineer AI to streamline life-saving antibiotic research

Researchers from the University of the Philippines Diliman—College of Science (UPD-CS) have engineered an artificial intelligence (AI) tool that can advance antibiotic findings, specifically on antibacterial peptides.

Key chemists Remmer Salas, Dr. Portia Mahal Sabido, and Dr. Ricky Nellas from UPD-CS Institute of Chemistry developed the instrument, dubbed as ISCAPE (Interpretable Support Vector Classifier of Antibacterial Activity of Peptides against Escherichia coli). 

ISCAPE acts as a computational biological tool, governed by AI mechanics, allowing researchers to predict the probability that a peptide inhibits the growth of E. coli bacteria.

In comparison to conventional methods that are time-consuming and complex, ISCAPE only requires a Simplified Molecular Input Line-Entry System (SMILES) string to evaluate candidate molecules, thus simplifying the process.

SMILES is a convenient method to read chemical structures in a concise format via line notation, essentially acting as a linguistic construct for chemistry-related research to which the chemists of this applied for their AI tool.

“ISCAPE helps address antimicrobial resistance by accelerating early-stage screening through data-driven peptide design,” Salas said, according to an article published by the UPD-CS.

Furthermore, it reduces the need for trial-and-error experiments and allows researchers to design better peptides—long protein chains—with efficiency and caution.

However, despite the cost-effectiveness and simplicity of the engineered innovation, real-life experiments conducted in laboratories (in vitro) or on whole biological organisms (in vivo) are more reliable in predicting antibacterial activity.

“Applying ISCAPE to other biological targets requires well-curated datasets with experimentally validated activity,” Salas added.

As such, the need for validated data through laboratory experiments is also critical, serving as a complement to ISCAPE or other biological predictive tools.

The chemists published ISCAPE on a web server and their paper in the Journal of Molecular Graphics and Modelling, being open to the public to view and analyze.