AI-powered literature mining reveals the therapeutic significance of GLP-1 receptor: Simulation of natural agonist candidates based on molecular dynamics.

Glucagon-like peptide-1 (GLP-1), a pivotal incretin hormone modulating glycemic homeostasis, has emerged as a clinically validated target for the treatment of type 2 diabetes and obesity. In this study, we present a comprehensive AI-integrated drug discovery pipeline that leverages BioBERT-based biomedical text mining to delineate the therapeutic landscape of GLP-1 receptor agonism systematically. Subsequent high-throughput virtual screening (HTVS) of a curated natural product library identified structurally diverse candidate ligands. A machine-learning-guided ADMET profiling algorithm was employed to prioritize compounds with optimal pharmacokinetic and safety characteristics. Top-ranked molecules were subjected to extensive molecular dynamics (MD) simulations using the GROMACS platform, enabling quantitative evaluation of structural stability, dynamic behavior, and receptor-ligand interaction persistence. Molecular docking analyses demonstrated robust binding affinities (ΔG: -11.3 to -8.7 kcal/mol), while MM-PBSA free energy estimations (ΔG<-30 kcal/mol) corroborated the thermodynamic favorability of binding. Among the screened entities, five lead candidates-CNP0244222.1, CNP0186692.11, CNP0361941.2, CNP0547477.1, and CNP0258197.2-consistently exhibited superior ADMET scores (>0.67), stable interaction trajectories, and enthalpically favorable profiles. This integrative, AI-augmented computational framework demonstrates substantial potential to accelerate the rational design and preclinical advancement of GLP-1-targeted therapeutics.