Supplementary Screening Dataset: Molecular Docking, Lipinski Filtering, ADMET, and ProTox Profiles of Phytochemical Candidates Against PirA/PirB Toxins

Description

This dataset contains the full computational screening data supporting the manuscript, covering molecular docking, Lipinski Rule of Five (Ro5) filtering, protein-ligand 2D interaction diagrams, ADME profiling, and ProTox-3.0 toxicity profiling of phytochemical candidates screened against the PirA (PDB: 3X0T) and PirB (PDB: 3X0U) toxin subunits of Vibrio parahaemolyticus.

Steps to reproduce

1. Phytochemical structures were compiled from literature review and retrieved as 3D SDF files from PubChem (https://pubchem.ncbi.nlm.nih.gov/). Crystal structures of PirA and PirB toxins (PDB: 3X0T, 3X0U) were obtained from the Protein Data Bank (https://www.rcsb.org/). 2. Ligands and proteins were prepared and converted to PDBQT format using PyRx v0.8. Active-site pockets were identified using BIOVIA Discovery Studio Visualizer v21.1.0.20298, and grid boxes were defined accordingly for each protein. 3. Virtual screening was performed using AutoDock Vina within PyRx v0.8, docking each compound against PirA and PirB independently. Binding affinities (kcal/mol) were recorded for all compounds and ranked in ascending order (most negative = most favorable), yielding the PirA_Docking_Scores and PirB_Docking_Scores sheets. 4. Each compound's best-scoring pose was evaluated against Lipinski's Rule of Five (H-bond donors, H-bond acceptors, molecular weight, XLogP) and assigned a Pass/Fail call, producing the PirA_LRo5 and PirB_LRo5 sheets. 5. For Ro5-Pass compounds, 2D protein-ligand interaction diagrams were generated in BIOVIA Discovery Studio Visualizer and visually inspected to apply an interaction-based filter: PirA: compounds were retained if they formed at least one bond (of any type) with the PirA active site, since overall interaction frequency against PirA was low across the library. This produced 12 candidates (Top 20% from PirA LRo5 Passed). PirB: compounds were excluded if their interaction diagram showed any unfavorable bond/contact, since interaction frequency against PirB was substantially higher and denser across the library, allowing a stricter criterion. This produced 42 candidates (Top 20% from PirB LRo5 Passed). 6. The retained candidates for each protein were submitted to SwissADME (www.swissadme.ch) to generate physicochemical and pharmacokinetic (ADME) profiles. 7. The same candidates were submitted to the ProTox 3.0 webserver (http://tox.charite.de/protox3/) to predict hepatotoxicity, carcinogenicity, immunotoxicity, mutagenicity, cytotoxicity, and ecotoxicity, using a rat model as the predictive standard. 8. Combining docking affinity, ADME suitability, and toxicity profile, the top three PirA candidates and top three PirB candidates were shortlisted (reported at the bottom of the PirA_ADME__Protox_Profile sheet, and in the corresponding manuscript tables for PirB) for downstream 100 ns molecular dynamics simulation (GROMACS 2024, CHARMM36m force field), reported in the main manuscript.

Institutions

• University of Dhaka

  Dhaka Division, Dhaka

Categories

Licence