UCARE Chem Suite and uCAREChemSuiteCLI: An online tool and a standalone R package for the prediction of drug resistome

Abstract

Background: In the era of antibiotic resistance, the prediction of bacterial resistome profiles, likely to cause resistance against novel drugs is of utmost importance. Despite this, to the best of our knowledge, no tool exists for the same. Therefore, with the rationale that drugs with similar structures have similar resistome profiles, two algorithms were developed viz., the deterministic model and the stochastic model to predict bacterial resistome for uncharacterized, but potential chemical structures. Methods and materials: Two separate databases were developed for Escherichia coli and Pseudomonas aeruginosa through exhaustive manual curation of published literature. Prediction of drug classes and bacterial resistome was carried out by the rules-based algorithm in the deterministic model, whereas the nearest neighbor algorithm was utilized in the stochastic model. Results: The models have been implemented in both standalone R package and an online server i.e., uCAREChemSuiteCLI and uCARE Chem Suite. In addition to resistome prediction, the online version of the suite enables the user to visualize the chemical structure, classify compound in predefined 19 drug classes, perform pairwise alignment and cluster with database compounds using graphical user interface. uCARE Chem Suite can be browsed from http://ucare.chem.e-bioinformatics.net, whereas uCAREChemSuiteCLI can be installed from CRAN (https://cran.r-project.org/package=uCAREChemSuiteCLI) and GitHub (https://github.com/sauravbsaha/uCAREChemSuiteCLI). Conclusion: We envisage that the said tool will enable the acceleration of the traditional drug discovery pipeline.