Machine learning-powered antibiotics phenotypic drug discovery

Abstract

Identification of novel antibiotics remains a major challenge for drug discovery. The present study explores use of phenotypic readouts beyond classical antibacterial growth inhibition adopting a combined multiparametric high content screening and genomic approach. Deployment of the semi-automated bacterial phenotypic fingerprint (BPF) profiling platform in conjunction with a machine learning-powered dataset analysis, effectively allowed us to narrow down, compare and predict compound mode of action (MoA). The method identifies weak antibacterial hits allowing full exploitation of low potency hits frequently discovered by routine antibacterial screening. We demonstrate that BPF classification tool can be successfully used to guide chemical structure activity relationship optimization, enabling antibiotic development and that this approach can be fruitfully applied across species. The BPF classification tool could be potentially applied in primary screening, effectively enabling identification of novel antibacterial compound hits and differentiating their MoA, hence widening the known antibacterial chemical space of existing pharmaceutical compound libraries. More generally, beyond the specific objective of the present work, the proposed approach could be profitably applied to a broader range of diseases amenable to phenotypic drug discovery.

A novel therapeutic antibody screening method using bacterial high-content imaging reveals functional antibody binding phenotypes of Escherichia coli ST131

“…HCI has been predominantly applied to mammalian cells and tissue where the examined variables include cell and organelle shape, signal transduction, gene expression and metabolism. In addition to studying mammalian cells, HCI has also been used to study intracellular pathogens such as Mycobacterium tuberculosis 19 , 20 , and more recently individual bacteria under drug exposure 21 . HCI has now reached a point where populations of bacteria can be phenotyped at single cell resolution in high-throughput to enable the simultaneous screening of multiple isolates of diverse bacterial clades.…”

Use of high-content analysis and machine learning to characterize complex microbial samples via morphological analysis

“…al. published a comprehensive description of various machine-learning techniques applied to imaged bacteria samples treated with compounds for the purpose of discovering novel antibiotic compounds[27]. We have utilized a similar approach to identify morphological changes of three different microbes in response to two antibiotics with different modes of action.…”

Hit Triage and Validation in Phenotypic Screening: Considerations and Strategies

“…Following data normalization, relevant parameters or principal components can be plotted on a multidimensional space to assess the clusterng of data points for each compound. This analysis can be used to explore the relationship between compound profiles and to hypothesize MoA based on the assumption that the same MoA would produce the same phenotype (Young et al, 2008;Zoffmann et al, 2019).…”