TB-ML—a framework for comparing machine learning approaches to predict drug resistance ofMycobacterium tuberculosis

Abstract

Motivation Machine learning (ML) has shown impressive performance in predicting antimicrobial resistance (AMR) from sequence data, including for Mycobacterium tuberculosis, the causative agent of tuberculosis. However, current ML development and publication practices make it difficult for researchers and clinicians to use, test, or reproduce published models. Results We packaged a number of published and unpublished ML models for predicting AMR of M. tuberculosis into Docker containers. Similarly, the pipelines required for pre-processing genomic data into the formats required by the models were also packaged into separate containers. By following a minimal container I/O standard we ensured as much interoperability as possible. We also created a command-line application, TB-ML, which can be used to easily combine pre-processing and prediction containers into complete pipelines ready for predicting resistance from novel, raw data with a single command. As long as there is adherence to this minimal standard for the container interface, containers produced by researchers holding new models can likewise be included in these pipelines, making benchmark comparisons of different models simple and facilitating faster uptake in the clinic. Availability and implementation TB-ML contains a simple Docker API written in Python and is available at https://github.com/jodyphelan/tb-ml. Example Docker containers for resistance prediction and corresponding data pre-processing as well as a tutorial on how to create new containers for TB-ML are available at https://tb-ml.github.io/tb-ml-containers/.

Multifactorial Analysis for Predicting Drug Resistance in Tuberculosis: Integrated Approaches to Improve Treatment

“…The review of the state of the art reveals a significant gap in advanced studies on drug resistance, with minimal usage of ML, as indicated by the references [16] [14] [17]. In light of this issue, the significance of this paper is underscored by the development of a predictive model designed to forecast the clinical outcomes related to TB drug resistance.…”

Machine learning and phylogenetic analysis allow for predicting antibiotic resistance inM. tuberculosis

“…Machine learning has been used for detecting resistance in Mtb data as well, employing different ML methods and achieved area under ROC curve values up to 0.95 in a classification task for resistance towards selected drugs using features from 23 selected target genes known to be implicated in resistance development (22). A recent computational framework, TB-ML, provides implementations for different ML methods such are random forest, direct association and convolutional neural networks (23). Training datasets for these methods can be found in public resources, such as, for example, the PATRIC database (24).…”

Machine learning and phylogenetic analysis allow for predicting antibiotic resistance in M. tuberculosis

“…Other approaches for detecting resistance in Mtb employed different ML models and achieved area under ROC curve values up to 0.95 in a classification task for resistance towards selected drugs using features from 23 selected target genes known to be implicated in resistance development [26,27]. A recent computational framework, TB-ML, provides implementations for different ML methods such are random forest, direct association and convolutional neural networks [28]. Treesist-TB, a customized decision tree-based machine learning algorithm for predicting resistance in Mtb, aims to extract genomic variants which might have been missed because of overfitting problems of the standard machine learning algorithms [29].…”