A Bilingual Computational Workflow for Identifying Potential PLK1 Inhibitors in American Sign Language and English

Polo-like Kinase 1 (PLK1) plays essential roles in the S, G2, and M phases of the cell cycle, and its overexpression is frequently observed in multiple cancers, including breast cancer, where it contributes to genomic instability and dysregulated apoptosis. Unlike conventional ATP-competitive inhibitors that target the kinase domain, selective inhibition of PLK1's polo-box domain (PBD) offers a promising strategy to disrupt protein-ligand interactions critical for mitotic progression, thereby triggering apoptosis in cancer cells. However, the high structural similarity between PLK1 and its homologs (PLK2 and PLK3), which are vital for neurological function and the stress response, respectively, necessitates exceptional selectivity to avoid off-target effects. To address this challenge, the protocol entails a bilingual (American Sign Language and English) computational workflow that integrates virtual screening, structural clustering, protein-ligand docking, binding affinity prediction, ADMET-S profiling, and quantum mechanical (QM) stability analysis. Starting from the SuperNatural 3.0 natural product database, compounds were filtered using breast cancer relevance and drug-likeness criteria, clustered to ensure chemical diversity, and evaluated their interactions with PLK1-, PLK2-, and PLK3-PBD structures. While virtual docking and in silico ADMET-S assessments cannot definitively confirm selectivity or mechanism of action, this study generates testable hypotheses and prioritizes a focused set of natural-product-derived candidates for future molecular dynamics simulations, biochemical validation, or experimental screening.