A White-Box Machine Learning Approach for Revealing Antibiotic Mechanisms of Action

  title={A White-Box Machine Learning Approach for Revealing Antibiotic Mechanisms of Action},
  author={Jason H. Yang and Sarah N. Wright and Meagan Hamblin and Douglas McCloskey and Miguel A. Cadena Alcantar and Lars Schr{\"u}bbers and Allison J. Lopatkin and Sangeeta Satish and Amir Nili and Bernhard O. Palsson and Graham C. Walker and James J. Collins},

A mechanism-aware and multiomic machine-learning pipeline characterizes yeast cell growth

This study proposes and test a machine-learning approach that integrates large-scale gene expression profiles and mechanistic metabolic models, for characterizing cell growth and understanding its driving mechanisms in Saccharomyces cerevisiae, and proposes a multiview neural network using fluxomic and transcriptomic data.

A flux-based machine learning model to simulate the impact of pathogen metabolic heterogeneity on drug interactions

A flexible machine-learning framework that utilizes diverse data types to effectively search through the large design space of both sequential and simultaneous combination therapies across hundreds of simulated growth conditions and pathogen metabolic states can serve as a useful guide for the selection of robustly synergistic drug combinations.

Decomposition of transcriptional responses provides insights into differential antibiotic susceptibility

It is shown that complex transcriptional changes induced by different media or antibiotic treatment can be traced back to a few key regulators, and fundamental shifts in respiration and iron availability that may explain media-dependent differential susceptibility to antibiotics are revealed.

Machine Learning of Bacterial Transcriptomes Reveals Responses Underlying Differential Antibiotic Susceptibility

This work investigates the interrelationship between antibiotic susceptibility and medium composition in Escherichia coli K-12 MG1655 as contextualized through machine learning of transcriptomics data and identified medium-dependent responses in key regulators of bacterial iron uptake and respiratory activity.

Predictive signatures of 19 antibiotics-induced Escherichia coli proteomes.

A comprehensive reference map of proteomic signatures of Escherichia coli under challenge of 19 individual antibiotics is presented, using label-free quantitative proteomics to derive a panel of 14 proteins that can be used to classify the antibiotics into different MOAs with nearly 100% accuracy.

Integrating genome-scale metabolic modelling and transfer learning for human gene regulatory network reconstruction.

A novel method for the reconstruction of the human gene regulatory network is proposed, based on a transfer learning strategy that synergically exploits information from human and mouse, conveyed by gene-related metabolic features generated in-silico from gene expression data.

A Deep Learning Approach to Antibiotic Discovery

Empowering systems-guided drug target discovery with metabolic and structural analysis

A systems-guided hierarchic workflow that utilizes metabolic and structural analysis to narrow in on specific targets suggested by statistical and machine learning analysis of metabolomics data is developed.

A genome-wide atlas of antibiotic susceptibility targets and pathways to tolerance

A genome-wide atlas of drug susceptibility determinants is built and a genetic interaction network that connects cellular processes and genes of unknown function is generated, which it is shown can be used as therapeutic targets.



Biological Machine Learning Combined with Campylobacter Population Genomics Reveals Virulence Gene Allelic Variants Cause Disease

The capability of machine learning coupled with GWAS and population genomics to simultaneously identify and rank alleles to define their role in infectious disease mechanisms is defined.

Using deep learning to model the hierarchical structure and function of a cell

DCell, a VNN embedded in the hierarchical structure of 2,526 subsystems comprising a eukaryotic cell, provides a foundation for decoding the genetics of disease, drug resistance and synthetic life.

Next-Generation Machine Learning for Biological Networks

How antibiotics kill bacteria: from targets to networks

The multilayered effects of drug–target interactions, including the essential cellular processes that are inhibited by bactericidal antibiotics and the associated cellular response mechanisms that contribute to killing are discussed.

Antibiotic efficacy-context matters.

Recon3D: A Resource Enabling A Three-Dimensional View of Gene Variation in Human Metabolism

Recon3D is presented, a computational resource that includes three-dimensional metabolite and protein structure data and enables integrated analyses of metabolic functions in humans and is used to functionally characterize mutations associated with disease, and identify metabolic response signatures that are caused by exposure to certain drugs.

Antibiotics induce redox-related physiological alterations as part of their lethality.

This work provides direct evidence that, downstream of their target-specific interactions, bactericidal antibiotics induce complex redox alterations that contribute to cellular damage and death, thus supporting an evolving, expanded model of antibiotic lethality.