A comprehensive automated computer-aided discovery pipeline from genomes to hit molecules

Abstract

Big data generation through sequencing of genomes and proteomes has led to over thousands of whole genomes and millions of protein sequences. However, utilization of this data to generate drug-like molecules for curing diseases remains a challenge. We propose here, Dhanvantari, a comprehensive software suite which automates the journey from genomes to hit molecules via its various modules such as i) gene finding, ii) computational structural study of target proteins and iii) virtual screening/identification of hit molecules for computer aided drug discovery. The pipeline has five possible entry points. Validation of the protocol is performed on 10 major life-threatening diseases reported by WHO covering 33 different protein targets and 111 FDA approved drugs. Three complete case studies from genomes to hits are also performed using the pipeline where the reported FDA drugs against the pathogen have been successfully recovered. The proposed software suite promises to deliver some new opportunities and insights into “genome-based” drug discovery along with the classical structure-based approaches to discover novel potential drug-like molecules. The entire protocol requires ~6–12 h. Individual steps, however, get completed within a few minutes. The pipeline can be freely accessed at http://www.scfbio-iitd.res.in/software/dhanvantari_new/Home.html.