Predicting drug-microbiome interactions with machine learning

:The microbiota of the gastrointestinal tract (GIT) is an extremely diverse community of microorganisms, and their collective genomes (microbiome) provide a vast arsenal of biological activities, particularly enzymatic ones, which are far from being fully elucidated. The study of the microbiota (and the microbiome) is receiving great interest from the biomedical community because it carries the potential to improve risk prediction models, refine primary and secondary prevention efforts, and also design more appropriate and personalized therapies, including pharmacological ones. A growing body of evidence, although sometimes impaired by the limited number of subjects involved in the studies, suggests that GIT dysbiosis, that is, the altered microbial composition, has an important role in causing and/or worsening cardiovascular disease (CVD). Bacterial translocation and the alteration of levels of microbe-derived metabolites can thus be important to monitor and modulate because they may lead to initiation and progression of CVD and to its establishment as chronic state. We hereby aim to provide readers with details on available resources and experimental approaches that are used in this fascinating field of biomedical research and on some novelties on the impact of GIT microbiota on CVD.

show abstract

“…138 Predicting drug-microbiome interactions with machine learning is also a promising advancement recently included to the field. 166…”

Section: Concentrations Of H 2 S and Quantification Methodsmentioning

confidence: 99%

Impact of the Gastrointestinal Tract Microbiota on Cardiovascular Health and Pathophysiology

Akyıldız

Biondi-Zoccai

Biase

2022

Journal of Cardiovascular Pharmacology

View full text Add to dashboard Cite

show abstract

“…Artificial intelligence (AI) based on machine learning (ML) has been increasingly used in different areas of knowledge. For example, ML techniques and algorithms allow a new analysis alternative and have accelerated discoveries of materials and formulations in the pharmaceutical field [37][38][39][40][41][42][43][44][45][46][47].…”

Section: Introductionmentioning

confidence: 99%

New Machine Learning Approach for the Optimization of Nano-Hybrid Formulations

et al. 2022

View full text Add to dashboard Cite

Nano-hybrid systems are products of interactions between organic and inorganic materials designed and planned to develop drug delivery platforms that can be self-assembled. Poloxamine, commercially available as Tetronic®, is formed by blocks of copolymers consisting of poly (ethylene oxide) (PEO) and poly (propylene oxide) (PPO) units arranged in a four-armed star shape. Structurally, Tetronics are similar to Pluronics®, with an additional feature as they are also pH-dependent due to their central ethylenediamine unit. Laponite is a synthetic clay arranged in the form of discs with a diameter of approximately 25 nm and a thickness of 1 nm. Both compounds are biocompatible and considered as candidates for the formation of carrier systems. The objective is to explore associations between a Tetronic (T1304) and LAP (Laponite) at concentrations of 1–20% (w/w) and 0–3% (w/w), respectively. Response surface methodology (RMS) and two types of machine learning (multilayer perceptron (MLP) and support vector machine (SVM)) were used to evaluate the physical behavior of the systems and the β-Lapachone (β-Lap) solubility in the systems. β-Lap (model drug with low solubility in water) has antiviral, antiparasitic, antitumor, and anti-inflammatory properties. The results show an adequate machine learning approach to predict the physical behavior of nanocarrier systems with and without the presence of LAP. Additionally, the analysis performed with SVM showed better results (R2 > 0.97) in terms of data adjustment in the evaluation of β-Lap solubility. Furthermore, this work presents a new methodology for classifying phase behavior using ML. The new methodology allows the creation of a phase behavior surface for different concentrations of T1304 and LAP at different pHs and temperatures. The machine learning strategies used were excellent in assisting in the optimized development of new nano-hybrid platforms.

show abstract

“…Moreover, it is desirable to minimise the use of animals, and sourcing either human faecal or intestinal fluid samples can be difficult and expensive. For this reason, in silico methods of predicting drugs' microbiome depletion hold significant potential [32][33][34][35][36].…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Predicts Drug Metabolism and Bioaccumulation by Intestinal Microbiota

et al. 2021

Self Cite

View full text Add to dashboard Cite

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.

show abstract

Predicting drug-microbiome interactions with machine learning

Cited by 58 publications

References 119 publications

Impact of the Gastrointestinal Tract Microbiota on Cardiovascular Health and Pathophysiology

Impact of the Gastrointestinal Tract Microbiota on Cardiovascular Health and Pathophysiology

New Machine Learning Approach for the Optimization of Nano-Hybrid Formulations

Machine Learning Predicts Drug Metabolism and Bioaccumulation by Intestinal Microbiota

Contact Info

Product

Resources

About