Abstract
Tuberculosis (TB) is caused by the bacterium Mycobacterium tuberculosis (Mtb) and is globally known for its high mortality rate due to its active pathogenic properties, which often result in inflammatory responses along the tissues they localize, causing necrosis of surrounding cells and even organ failure. Based on the organ and tissue-specific infection caused by Mtb, TB is classified into two categories: pulmonary TB (PTB) and extrapulmonary TB (EPTB), of which the latter is known for its infecting capability across various organs like the brain, liver, lymph nodes, etc. Hence, to analyze the biological pathways and genes enriched within the progression of PTB into EPTB, along with the detailed analysis of biological enrichment and relationships shared among the various categories of EPTB, we present a One Health model that utilizes series of computational clustering methodologies to analyze the similar genes and biological pathways that correspond to the disease types of EPTB, along with the development of knowledge graphs to visualize the biological interactions among various chemicals and genes that contribute to therapeutic mechanisms of action to treat the disease symptoms and correspond to both PTB and EPTB, respectively. Ultimately, we present a Read Across QSAR (RASAR) model that uses machine learning approaches to predict the therapeutic nature of novel chemical candidates by assessing their morphological descriptors. Henceforth, by deploying a series of end-to-end computational strategies, we successfully analyze the complex biological data associated with the pathogenesis of EPTB and establish their significance with PTB, along with screening out potential biological pathways, genes, and chemicals that specifically correspond to the development of TB within patients.
Supplementary materials
Title
Supplimentary material
Description
Supplementary data
Actions