Bayesian Modeling and Intrabacterial Drug Metabolism Applied to Drug-Resistant Staphylococcus aureus

Abstract

We present the application of Bayesian modeling to identify chemical tools and/or drug discovery entities pertinent to drug-resistant Staphylococcus aureus infections. The quinoline JSF-3151 was predicted by modeling and then empirically demonstrated to be active against in vitro cultured clinical methicillin- and vancomycin-resistant strains while also exhibiting efficacy in a mouse peritonitis model of methicillin-resistant S. aureus infection. We highlight the utility of an intrabacterial drug metabolism (IBDM) approach to probe the mechanism by which JSF-3151 is transformed within the bacteria. We also identify and then validate two mechanisms of resistance in S. aureus: one mechanism involves increased expression of a lipocalin protein, and the other arises from the loss of function of an azoreductase. The computational and experimental approaches, discovery of an antibacterial agent, and elucidated resistance mechanisms collectively hold promise to advance our understanding of therapeutic regimens for drug-resistant S. aureus.

Random Forest Model Predictions Afford Dual-Stage Antimalarial Agents

“…The model performed well in our opinion, making correct predictions among labeled active (26%) and inactive (95%) candidates as based on single-concentration assays. This performance is consistent with our experience with our in vitro efficacy models for a range of bacteria. ,,, The machine learning approach offers hit rates considerably better than typical empirical screening (≤0.5%) while demonstrating a significant ability to predict inactive compounds. For resource-limited screening efforts, especially for parasitic diseases such as malaria, we view this as a meaningful savings in time and cost to identify active compounds for downstream study.…”

“…The efforts herein constitute, to the best of our knowledge, the first reported utilization of machine learning methods to predict in vitro antimalarial liver-stage efficacy. It builds on our previous efforts predicting antibacterial in vitro efficacy ,,, and molecular properties − critical to the attainment of high-quality chemical tools and drug discovery agents. Leveraging a data set of 5972 compounds, a random forest model was constructed, validated, and then utilized to predict the liver-stage efficacy of drug-like small molecules from a ChemDiv diversity library.…”

Rickettsia Aglow: A Fluorescence Assay and Machine Learning Model to Identify Inhibitors of Intracellular Infection

“…The data from the MicroSource compound library screen were utilized to construct a Bayesian model. This methodology has been previously employed in our laboratory with growth inhibition screening data with other bacteria. − Model construction relied on the practical definition of an active compound at a 10 μM concentration achieving ≥50% growth inhibition in the Vero cell infection screen and exhibiting <25% growth inhibition of uninfected Vero cells. Model construction involved the examination of eight previously used calculated molecular features (ALogP, molecular weight, number of rotatable bonds, number of aromatic rings, total number of all types of rings, number of hydrogen-bond donors, number of hydrogen-bond acceptors, and the molecular fractional polar surface area) and either FCFP6 or ECFP6 fingerprints .…”

“…Early contributions in this realm include reports from Prathipati and us that focused primarily on Mycobacterium tuberculosis and Bayesian models. 22,52,53 The reports in recent years have expanded to include other bacteria, including but not limited to Staphylococcus aureus, 23 Neisseria gonorrhoeae, 24 and Escherichia coli. 54,55 The machine learning methods leveraged in these publications now include approaches other than Bayesian models, 23,24 such as support vector machines, 56 decision trees, 55,56 and directed message passaging deep neural networks.…”

“…For these reasons, we commenced a program to search for novel therapeutic strategies to inhibit rickettsial infections, with a specific focus on the model organism Rickettsia canadensis. , We chose an approach that would commence with the design of a high-throughput screen for the growth inhibition of this bacterium within the context of an intracellular infection. We subsequently leveraged the resulting data set and our experience with machine learning models for the growth inhibition of Mycobacterium tuberculosis , Staphylococcus aureus, and Neisseria gonorrhoeae to train a Bayesian model relevant to intracellular rickettsial infection. Herein, we describe assay design and validation, use of the resulting data to train the model and predict candidate hit compounds, and exploration of the empirically validated hit compounds that point toward future translational studies.…”