Artificial Intelligence and Machine learning based prediction of resistant and susceptible mutations in Mycobacterium tuberculosis

Abstract

Tuberculosis (TB), an infectious disease caused by Mycobacterium tuberculosis (M.tb), causes highest number of deaths globally for any bacterial disease necessitating novel diagnosis and treatment strategies. High-throughput sequencing methods generate a large amount of data which could be exploited in determining multi-drug resistant (MDR-TB) associated mutations. The present work is a computational framework that uses artificial intelligence (AI) based machine learning (ML) approaches for predicting resistance in the genes rpoB, inhA, katG, pncA, gyrA and gyrB for the drugs rifampicin, isoniazid, pyrazinamide and fluoroquinolones. The single nucleotide variations were represented by several sequence and structural features that indicate the influence of mutations on the target protein coded by each gene. We used ML algorithms-naïve bayes, k nearest neighbor, support vector machine, and artificial neural network, to build the prediction models. The classification models had an average accuracy of 85% across all examined genes and were evaluated on an external unseen dataset to demonstrate their application. Further, molecular docking and molecular dynamics simulations were performed for wild type and predicted resistance causing mutant protein and anti-TB drug complexes to study their impact on the conformation of proteins to confirm the observed phenotype.

Insights Into Mutations Induced Conformational Changes and Rearrangement of Fe2+ Ion in pncA Gene of Mycobacterium tuberculosis to Decipher the Mechanism of Resistance to Pyrazinamide

“…These patterns suggest that the internal dynamics of each system is impacted by the mutation and eventually contributed to the PZA resistance. This notion is also supported by previous study ( Jamal et al, 2020 ; Karmakar et al, 2020 ). The lowest energy minima conformation from each trajectory was extracted and compared with the native state that reported significant variations in Fe 2+ binding and β–stands 2 specifically also confirmed by published literature ( Khan et al, 2019c ; Junaid et al, 2020 ).…”

Mycobacterial drug discovery

“…Machine learning can also be applied for mode of action determination, based on known drug-target interactions, 77 as well as for the predication of resistance conferring mutations. 78 The machine learning models are developed with the aim of increasing the efficiency of drug discovery.…”

Prediction of antibiotic resistant strains of bacteria from their beta-lactamases protein

“…Thus, we have designed a model for predicting beta-lactamase variants that make ceftazidime sensitive or resistant to a bacterial strain. We used state of the arts techniques mainly based on machine learning techniques to develop prediction models [15]. This will help in predicting ceftazidime resistance/susceptibility towards beta-lactamase carrying bacterial species that could emerge in the near future.…”