Machine Learning Predicts 1-[(4-Fluorophenyl) Methyl] Indole-2,3- Dione as Drug Lead for Peptide Deformylase in Plasmodium falciparum

Abstract

Malaria continues to be a serious illness for world health. Deadly parasites that cause malaria infect female Anopheles mosquitoes and bite humans are the disease's vector. Since Plasmodium falciparum is the deadliest of the five Plasmodium species and research indicates that medicines for malaria are increasingly revealing drug-resistance mechanisms, it is imperative that new and effective medications be developed. This project aims to identify and produce a novel therapeutic lead against a P. falciparum target that has been validated. Target2Scan, a programmed tool, was employed in this investigation to find potential therapeutic targets for P. falciparum peptide deformylase. After the tool received the target signature, it ran a BLAST methodology to find targets that resembled peptide deformylase. To ascertain the binding affinities of protein-ligand complexes, molecular docking was performed using PyRx and CBDock. Of the 12,561 compounds produced, 11,304 were used as the training set and 1,256 as the test set. Six ligands were produced by machine learning as potential therapeutic leads, and using molecular docking, 1-[(4-fluorophenyl) methyl] indole-2,3-dione showed the greatest biding effect on peptide deformylase when compared to 5-chloro-1-(2-phenylethyl) indole-2,3-dione. The higher binding affinity of 1-[(4-fluorophenyl) methyl] indole-2,3-dione over 5-chloro-1-(2-phenylethyl) indole-2,3-dione and other ligands on peptide deformylase suggests that it has the ability to suppress P. falciparum pepide deformylase activity.