Computation of host-guest binding free energies with a new quantum mechanics based mining minima algorithm

Area of Science:

• Computational Chemistry
• Molecular Modeling
• Supramolecular Chemistry

Background:

• Accurate calculation of noncovalent binding free energies is crucial for understanding host-guest interactions.
• Existing methods often struggle with the computational cost and accuracy for diverse host-guest systems.

Purpose of the Study:

• To develop and validate the QM-VM2 method for efficient and accurate calculation of host-guest binding free energies.
• To assess the impact of different quantum mechanical levels and entropy on prediction accuracy.

Main Methods:

• QM-VM2 combines semi-empirical quantum mechanics (SEQM) with molecular mechanics (MM) conformational search.
• Employs a progressive geometry optimization scheme for enhanced sampling efficiency.
• Utilizes post-processing ab initio quantum mechanics (QM) calculations and includes configurational entropy.

Main Results:

• SEQM-VM2 alone showed poor correlation for all charges but good correlation (R²=0.8) for +1 charges.
• Ab initio QM post-processing significantly improved correlation for the full set (R²=0.6-0.7) and +1 set (R²=0.9-1.0).
• Inclusion of configurational entropy (ΔE - TΔS) was essential for good experimental correlation.

Conclusions:

• QM-VM2, with appropriate QM post-processing and entropy, provides accurate binding free energy predictions for host-guest systems.
• The method shows particular promise for systems with +1 formal charges.
• An empirical interpolation scheme can further improve accuracy across a range of system charges.