A program to automate the discovery of drugs for West Nile and Dengue virus-programmatic screening of over a billion compounds on PubChem, generation of drug leads and automated in silico modelling.

Our work is composed of a python program for programmatic data mining of PubChem to collect data to implement a machine learning-based AutoQSAR algorithm to generate drug leads for the flaviviruses-Dengue and West Nile. The drug leads generated by the program are fed as programmatic inputs to AutoDock Vina package for automated in silico modelling of interaction between the compounds generated as drug leads by the program and the chosen Dengue and West Nile drug target methyltransferase, whose inhibition leads to the control of viral replication. The machine learning-based AutoQSAR algorithm involves feature selection, QSAR modelling, validation and prediction. The drug leads generated, each time the program is run, are reflective of the constantly growing PubChem database which is an important dynamic feature of the program which facilitates fast and dynamic drug lead generation against the West Nile and Dengue viruses. The program prints out the top drug leads after screening PubChem library which is over a billion compounds. The interaction of top drug lead compounds generated by the program and drug targets of West Nile and Dengue virus was modelled in an automated way through the tool. The results are stored in the working folder of the user. Thus, our program ushers in a new age of automatic ease in the virtual drug screening and drug identification through programmatic data mining of chemical data libraries and drug lead generation through machine learning-based AutoQSAR algorithm and an automated in silico modelling run through the program to study the interaction between the drug lead compounds and the drug target protein of West Nile and Dengue virus. The program is hosted, maintained and supported at the GitHub repository link given below Communicated by Ramaswamy H. Sarma.