Students from Springwest Academy with Orbyts Fellow Marc Amil from London Interdisciplinary Biosciences Consortium used artificial intelligence (AI) and language models like ChatGPT to discover novel antimicrobial peptides (AMPs) as potential solutions to antibiotic resistance! Antibiotic resistance is a growing global health crisis, threatening to cause millions of deaths annually.

This study utilised a dataset of AMPs and non-AMP peptides to train and test an AI model. The model demonstrated improved specificity and accuracy compared to previous machine learning models in classifying AMPs.
Interestingly, the students found that AMP residues tend to have higher hydrophobicity and isoelectric points compared to non-AMP residues. These findings align with previous studies highlighting the importance of these properties in AMP activity.

The success of this AI model in accurately identifying AMPs suggests its potential to accelerate the discovery of new antimicrobial agents. By leveraging the capabilities of language models, researchers hope to generate novel AMPs that can effectively combat antibiotic-resistant bacteria!