Computational identification of potential PAK1 inhibitors for anti-cancer therapy: an e-pharmacophore guided virtual screening study

Cancer remains a major global health challenge, with approximately 18 million new cases reported annually. Existing evidence highlights the PAK1 protein as a critical regulator of cancer progression, making it a promising therapeutic target. The PAK1 protein complexed with dibenzodiazepine was fetched from the PDB with the identifier 4ZLO. The structure was preprocessed through preparation and exposed to pharmacophore hypotheses on the Schrödinger suite programme, indicating key features of RRH. A multi-tiered docking-based screening workflow from the libraries of ZINC and Enamine databases identified five potential bioactive compounds: ZINC952869440, ZINC952869442, ENAMINE558, ENAMINE6304, and ENAMINE8429. The docking and MM/GBSA scores ranked from -5.02 to -8.34 kcal/mol and -46.10 to -50.41 kcal/mol. Remarkably, none of these candidates violated the rules of five, and the Qikprop parameters complied with pharmacokinetic suitability. The DFT analysis revealed energy gap scores ranged from -0.182 to -0.225 eV, indicating favourable electronic properties and stability of the ligands. Furthermore, molecular dynamics (MD) and essential dynamics (ED) studies validated the structural stability of the complexes. The secondary structure analysis indicated stable retention of α-helices and β-strands throughout the simulation. Moreover, the computational investigation identified potential PAK1 inhibitors that warrant further experimental testing and therapeutic development.