Software

The virtual screening of large numbers of compounds against target protein binding sites has become an integral component of drug discovery workflows. This screening is often done by computationally docking ligands into a protein binding site of interest, but this has the limitation that a large number of poses must be evaluated to obtain accurate estimates of protein-ligand binding affinity. In RASPD+, we introduced a fast prefiltering method for ligand prioritization that is based on a set of machine learning models and uses simple pose-invariant physicochemical descriptors of the ligands and the protein binding pocket. This method, Rapid Screening with Physicochemical Descriptors + machine learning (RASPD+), is trained on PDBbind data and achieves a regression performance better than for the original RASPD method and comparable to traditional scoring functions on a range of different test sets without the need for generating ligand poses. Additionally, we use RASPD+ to identify molecular features important for binding affinity and assess the ability of RASPD+ to enrich active molecules from decoys. This method is freely accessible on GitHub

I am one of the co-developers of Sanjeevini. This software has been developed to provide a computational guidance for lead-like molecules design. The user can upload a bimolecular (protein) target and a candidate drug. The software identifies the potential active sites, screen million molecule library and then docks and scores the candidates and returns structures of the candidate molecule bound to protein target together with binding free energies.

Xenobiotic metabolism can produce metabolites with physico-chemical and biological properties that differ substantially from those of the parent compound. Hence the prediction of the metabolic fate of small molecules is of critical importance to the development of safe and efficacious medicines, cosmetics, agrochemicals and in fact, any chemicals exposed to biological systems. Cytochrome P450s (CYPs), a superfamily of heme containing enzymes, are the major enzymes involved in drug metabolism. Binding modes of ligands with CYP decide the metabolic products. Predicting correct binding modes for a ligand with respect to any enzyme is a necessary step in structure-based drug-design. This methodology combines molecular docking to assess atomic accessibility to the heme iron of cytochrome P450 enzymes with molecular orbital (MO) calculations to evaluate the reactivity of the accessible atomic sites, aiming to predict the metabolic sites of a molecule.

This tool is a computationally fast protocol for identifying good candidates for any target protein from any molecule/million molecule database. A QSAR-type equation sets up the extent of complementarity of the physicochemical properties of the target protein and the candidate molecule and an estimate of the binding energy is generated. The most interesting feature of this methodology is that it takes only a fraction of a second for calculating the binding energy of any ligand without docking in the active site of the target protein as opposed to several minutes for regular docking and scoring methods, while the accuracy in sorting good candidates remains comparable to that of conventional techniques. An entire million compound library can be scanned against a specified target protein within a few minutes in a single core machine for identifying hit molecules.

This tool checks the Lipinski compliance of small organic molecules. Using this tool, one computes the physicochemical parameters of small molecules and based on the values, it is possible to distinguish between drug like and non-drug like molecules.

 The following physicochemical properties one can check

Molecular weight (range for drug-like molecule: <500 Da)

Lipophilicity (range for drug-like molecule: <5)

Hydrogen bond donors (range for drug-like molecule: <10)

hydrogen bond acceptors (range for drug-like molecule: <10)

Molar refractivity (range for drug-like molecule: 40-130 

This is a fingerprint (atom types up to 4-bonds path) based partial atomic charge derivation method of small organic molecule. In this protocol, a lookup table of 5302 atom types to cover the chemical space of small molecules which contains of C, H, O, N, S, P, F, Cl and Br elements together with their quantum mechanical RESP fit charges, has been created. The partial charge on any atom in a query molecule is then assigned by reference to the look-up table. The accuracies attained are comparable to more rigorous but time-consuming ESP derived method like, RESP.