Scalable Analysis of Dipole Moment Fluctuations for Characterizing Intermolecular Interactions and Structural Stability.

Accurately predicting protein-ligand interactions is essential in computational molecular biochemistry and in silico drug development. Monitoring changes in molecular dipole moments through molecular dynamics simulations provides valuable insights into dipole-dipole interactions, which are critical for understanding protein structure stability and predicting protein-ligand binding affinity. In this study, we propose a novel method to monitor changes in the interangle between dipole vectors of residue molecules within proteins and ligand molecules, aiming to evaluate the strength and consistency of interactions within the complex. Additionally, we extend the concept of positional root-mean-square fluctuation (RMSF), commonly used for protein structure stability analysis, to dipole moments, thus defining dipole moment RMSF. This enables us to analyze the stability of dipole moments for each residue within the protein and compare them across residues and between binding and non-binding complexes. Using the CRBP1-retinoic acid complex as our model system, we observed a significant difference in the interangle change of dipole moments for the key residue at the residue-level between the non-binding and binding complexes. Furthermore, we found that the dipole moment RMSF value of the non-binding complex was substantially larger than that of the binding complex, indicating greater dipole moment instability in the non-binding complex. Leveraging the concept of scalability inherent in the calculation of dipole moment vectors, we systematically expanded the residues within the protein's primary secondary structure. Our dipole moment analysis approach can provide valuable predictive insights into complex candidates, especially in situations where experimental comparisons are challenging.