Conformational Free-Energy Differences of Large Solvated Systems with the Focused Confinement Method

The focused confinement method (FCM) is a reaction coordinate-free simulation approach for the calculation of conformational free-energy differences in explicit solvent. The method uses reference states for the conformations of interest, partitions the solute into conformationally active and inactiv...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of chemical theory and computation 2020-08, Vol.16 (8), p.5163-5173
Hauptverfasser: Orndorff, Paul B, Le Phan, Sang T, Li, Ka Ho, van der Vaart, Arjan
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The focused confinement method (FCM) is a reaction coordinate-free simulation approach for the calculation of conformational free-energy differences in explicit solvent. The method uses reference states for the conformations of interest, partitions the solute into conformationally active and inactive regions, and requires the calculation of desolvation free energies of mixed harmonic–anharmonic states as part of its procedure. The reference states and partitioning affect the speed of convergence of FCM’s constituent simulations in opposing manners, but in the thermodynamic limit, they have no effect on calculated conformational free-energy differences. To aid fast convergence of large systems, a general procedure to quickly partition and construct reference states is introduced. With this, two sets of reference states and associated partitionings were constructed for the closed and open conformation of triosephosphate isomerase (TIM). Despite TIM’s size, highly converged desolvation free energies were readily obtained from standard free-energy perturbation simulations because the mixed harmonic–anharmonic states are heavily rigidified. FCM-calculated free-energy differences for loop closing matched the experimental value for both reference sets. The insensitivity to reference states and associated partitionings favors reference states that merely reflect main structural differences for which convergence is faster. The calculations demonstrate the accuracy and robustness of FCM for large systems.
ISSN:1549-9618
1549-9626
DOI:10.1021/acs.jctc.0c00403