CORRELATION BETWEEN MISSING LOOP PROXIMITY TO THE BINDING SITE AND GRID SCORE CHANGES AFTER MISSING LOOP REFINEMENT

Abstract

This study investigated whether missing loops and ligand distance is associated with grid score changes after refinement. Total of 23 protein and ligand complexes were selected from the Protein Data Bank based on resolution of ≤ 2.5 Å, presence of a co-crystallized ligand, and missing loop regions. The structures were classified into close, moderate, and far groups using UCSF Chimera 1.16. Missing loops were reconstructed using MODELLER 10.8, followed by molecular docking refinement using DOCK6. ΔGrid Score were calculated using Microsoft Excel, ligand-receptor interactions were visualized using BIOVIA Discovery Studio Visualizer 2025, and statistical analysis was performed using IBM SPSS Statistics. The close group showed the highest mean absolute grid score change at 18.7741 kcal/mol, followed by the moderate group at 2.2939 kcal/mol and the far group at 0.7025 kcal/mol. Spearman analysis showed a significant negative relationship between missing loop and ligand distance and absolute grid score change, with a correlation coefficient of -0.430 and a p-value of 0.040. These findings indicate that missing loops located closer to ligands tend to produce larger docking score changes after refinement, although score direction is also influenced by local interaction patterns.