Text Mining for Identification of Anti-Tubercular Compounds Present in Plants of India and In-Silico ADMET Prediction of Phytochemicals

Abstract

Tuberculosis (TB) is a highly contagious disease and a potent killer. It infects around one-third of the world’s population. TB is curable and preventable but extensive and unsupervised use of antibiotics is resulting in the case of drug resistance TB. Therapeutics for tuberculosis take a long time to combat TB and have side effects. Thus, the need for new drugs persists. From ancient times plant-derived chemotherapeutic agents have been used in the treatment of several diseases. Modern biological research depends on access to the biological information encoded in expert-curated databases. It is a difficult task for the biologist to manually curate all the knowledge of the rapidly growing biological literature. The computer-assisted curation helps the efficiency of the literature search through the text mining approach. The present study aims at the text mining of the Indian plants, showing Antitubercular properties, and investigate the pharmacokinetics of compounds present in these plants. PubTator, a web-based system was used for text mining. Total 66 works of literature were found carrying the search terms. Plants having anti-tubercular properties were identified from India by curating PubMed abstracts. The information of phytocompounds was extracted for developing a local library and ADMET analysis. The result of this study could be further used for the development of drugs against tuberculosis.