Synergizing Machine Learning and fluorescent biomolecules: A new era in sensing platforms

Abstract

Machine Learning (ML) algorithms offer significant advantages over traditional methods, enabling the identification of complex correlations and hidden patterns within data, which enhances efficiency, reduces costs, and improves decision-making. This article provides a comprehensive overview of recent advances in ML-assisted fluorescent peptide and protein-based sensors. Notably, a supervised ML-assisted peptide-based sensor has been developed for the identification of water-soluble polymers, improving environmental and industrial monitoring. ML-assisted sulfonamido-oxine (SOX)-labeled peptides facilitate the quantitation of mitogen-activated protein kinases, advancing sensitive biomarker analysis. An array-based detection system using green fluorescent protein conjugates enables high-throughput protein screening. A deep learning (DL)-assisted fluorophore-labeled peptide sensor array shows promise for non-invasive breast cancer diagnosis with high accuracy. Additionally, a ML-aided sensor array combining antimicrobial peptides and fluorescent proteins enables the discrimination of top clinical isolates, enhancing antimicrobial resistance diagnostics. These innovations in peptide sensor design and ML integration highlight their transformative impact in biological research, disease diagnostics, and environmental monitoring, offering improved sensitivity, selectivity, and performance. This review provides valuable insights for researchers and practitioners in the field of fluorescence-based sensing, ML, and their interdisciplinary applications.