Docking & ADMET for Protease

Python, Data Visualization, Data Analysis, Bioinformatics · I have an in-silico project that combines structure-based and ligand-based techniques. The goal is to identify promising small-molecule inhibitors for a specific protease, then evaluate their pharmacokinetic and toxicity profile and, finally, build a predictive QSAR model to prioritise leads.

First, I will supply the crystal structure (PDB) of the protease along with a starter ligand set. I need you to prepare the protein, generate the receptor grid, and run molecular docking for the compound library, ranking poses by binding affinity and key interactions. You may use AutoDock Vina, Schrödinger Glide, MOE, or any equivalent docking engine you are comfortable with, provided the workflow is reproducible.

Next comes ADME / ADMET profiling. Using tools such as SwissADME, ADMETlab, pkCSM, or QikProp, produce a concise report that flags oral bioavailability, metabolic liabilities, hERG risk, and any major red-flag toxicities for the top-scoring docked hits.

Finally, build a QSAR model that correlates molecular descriptors with the docking and ADMET results to highlight structure–activity trends and suggest modifications. You’re free to work in Python (scikit-learn, RDKit) or commercia