Tree-Based QSAR Model for Drug Repurposing in the Discovery of New Antibacterial Compounds against Escherichia coli

Abstract

Drug repurposing appears as an increasing popular tool in the search of new treatment options against bacteria. In this paper, a tree-based classification method using Linear Discriminant Analysis (LDA) and discrete indexes was used to create a QSAR (Quantitative Structure-Activity Relationship) model to predict antibacterial activity against Escherichia coli. The model consists on a hierarchical decision tree in which a discrete index is used to divide compounds into groups according to their values for said index in order to construct probability spaces. The second step consists in the calculation of a discriminant function which determines the prediction of the model. The model was used to screen the DrugBank database, identifying 134 drugs as possible antibacterial candidates. Out of these 134 drugs, 8 were antibacterial drugs, 67 were drugs approved for different pathologies and 55 were drugs in experimental stages. This methodology has proven to be a viable alternative to the traditional methods used to obtain prediction models based on LDA and its application provides interesting new drug candidates to be studied as repurposed antibacterial treatments. Furthermore, the topological indexes Nclass and Numhba have proven to have the ability to group active compounds effectively, which suggests a close relationship between them and the antibacterial activity of compounds against E. coli.

Potential Repurposed Drug Candidates for Tuberculosis Treatment: Progress and Update of Drugs Identified in Over a Decade

“…The process begins with the large pool of drugs (old/failed/investigational/FDA approved). Various computational or experimental repurposing approaches like molecular docking, 14 pharmacophore modeling, 15 and high throughput drug screening (HTDS) 16 are used to screen these drugs for their potential to inhibit M.tb survival. From such screening, the lead bioactive compounds are selected for further validations in preclinical and clinical trial studies to evaluate molecular pathophysiology, contraindications, dosage, and synergistic effects.…”

Virtual Combinatorial Chemistry and Pharmacological Screening: A Short Guide to Drug Design

“…They used this model to virtually screen a fluoroquinolone VCL, identifying 117 theoretically active molecules of which five were synthesized and three showed anti-MRSA activity comparable to that of ciprofloxacin. Similarly, Suay-Garcia et al developed a tree-based QSAR model based on quinolones that was applied to the DrugBank database to screen for active compounds against Escherichia coli [75]. The model identified 134 drugs with theoretical activity against E. coli of which eight were already commercialized as antibacterial drugs, 67 were approved for different pathologies, and 55 were drugs in experimental stages.…”

Machine Learning in Antibacterial Drug Design

“…In this review, a number of RF applications are presented, for small molecules, peptides ( Bhadra et al, 2018 ), natural product–based antibacterial design, and studying antibacterial drug resistance ( Dias et al, 2019 ; Maltarollo et al, 2019 ; Shi et al, 2020 ; Li et al, 2021 ; Wani et al, 2021 ). A good example of underlying supervised learning DT method was reported by Suay-Garcia et al (2020) . The authors created a QSAR model to predict antibacterial activity against E. coli .…”