Machine Learning Predicts Drug Metabolism and Bioaccumulation by Intestinal Microbiota

Abstract

Over 150 drugs are currently recognised as being susceptible to metabolism or bioaccumulation (together described as depletion) by gastrointestinal microorganisms; however, the true number is likely higher. Microbial drug depletion is often variable between and within individuals, depending on their unique composition of gut microbiota. Such variability can lead to significant differences in pharmacokinetics, which may be associated with dosing difficulties and lack of medication response. In this study, literature mining and unsupervised learning were used to curate a dataset of 455 drug–microbiota interactions. From this, 11 supervised learning models were developed that could predict drugs’ susceptibility to depletion by gut microbiota. The best model, a tuned extremely randomised trees classifier, achieved performance metrics of AUROC: 75.1% ± 6.8; weighted recall: 79.2% ± 3.9; balanced accuracy: 69.0% ± 4.6; and weighted precision: 80.2% ± 3.7 when validated on 91 drugs. This machine learning model is the first of its kind and provides a rapid, reliable, and resource-friendly tool for researchers and industry professionals to screen drugs for susceptibility to depletion by gut microbiota. The recognition of drug–microbiome interactions can support successful drug development and promote better formulations and dosage regimens for patients.

Colonic drug delivery: Formulating the next generation of colon-targeted therapeutics

“…The substantial metabolic capacity of the microbiome can alter the pharmacokinetics (PK) of drugs, and thus their therapeutic efficacy, via a plethora of mechanisms in a patient-specific manner [128,136,137]. The intestinal concentrations of over 150 small molecule drugs are currently known to be directly depleted by microbiota [138]. It has recently been demonstrated that gut microbiota can accumulate drugs as well as chemically transform them [139].…”

“…Elsewhere, an ANN predicted the size and drug loading efficiency of nanoparticles based on excipient composition, resulting in more accurate predictions than a DoE model applied to the same task [272]. ML has also been leveraged to predict drug-microbiome interactions [136,138,140]. As the amount of data available to research departments continues to increase, quantum computing may supersede classical computing for ML and other in silico techniques, due to its capacity to process datasets containing hundreds of thousands of datapoints much more efficiently [273,274].…”

Gut Microbiome: The Interplay of an “Invisible Organ” with Herbal Medicine and Its Derived Compounds in Chronic Metabolic Disorders

“…Only a small proportion of the interactions between natural drugs and GM have been elucidated, considering the huge contribution of bacterial metabolism in digestion [ 123 , 124 ]. A recent study suggested a novel solution: adopting machine learning to predict drugs’ metabolism by GM [ 125 ]. Although the model used in the study predicted the depletion of drugs by gut microbial metabolism and did not suggest any consequent metabolites, it is worth exploring the possibilities of computational analysis in this field.…”

Machine learning using multi-modal data predicts the production of selective laser sintered 3D printed drug products

“…In healthcare, ML is used in clinical trials, diagnostics and surgery (Giorgio et al, 2022;Halamka et al, 2022;Myszczynska et al, 2020;Shah et al, 2019;Zame et al, 2020). In pharmaceutics, ML models have been applied to model drug-food interactions, drug-microbiome interactions, and formulation development (Gavins et al, 2022;McCoubrey et al, 2021;Wang et al, 2022). For 3D printing medicines, ML has been demonstrated to predict printability, drug release rate, and accelerating quality control (Elbadawi et al, 2020a;Muñiz Castro et al, 2021;O'Reilly et al, 2021).…”