OMx-D: semiempirical methods with orthogonalization and dispersion corrections. Implementation and biochemical...

Area of Science:

• Computational chemistry
• Biomolecular modeling

Background:

• Standard semiempirical methods like AM1 have limitations in describing biochemical systems.
• The OMx family (OM1, OM2, OM3) offers improved accuracy for biochemical systems.
• Accurate modeling of non-covalent interactions is crucial in biomolecular studies.

Purpose of the Study:

• To investigate the benefits of augmenting OMx semiempirical methods with empirical dispersion terms (OMx-D).
• To evaluate the performance of OMx-D methods for non-covalent interactions and binding energies.
• To assess the utility of OMx-D in quantum mechanical/molecular mechanical (QM/MM) simulations of biomolecules.

Main Methods:

• Augmentation of OMx (OM1, OM2, OM3) methods with an empirical dispersion term (OMx-D).
• Evaluation using the JSCH-2005 database for complex binding energies.
• Application in quantum mechanics (QM) and QM/molecular mechanics (QM/MM) calculations for antibody-hapten binding.

Main Results:

• OMx-D methods significantly improve the description of non-covalent interactions.
• Mean unsigned errors for binding energy were 1.41 kcal/mol (OM2-D) and 1.31 kcal/mol (OM3-D).
• OM3-D demonstrated superior performance compared to AM1 and OM3 in QM/MM calculations of antibody-hapten binding.

Conclusions:

• Augmented OMx-D methods provide enhanced accuracy for non-covalent interactions in biochemical systems.
• OMx-D methods are suitable for QM/MM studies, particularly during system preparation and equilibration.
• The OMx-D approach represents a valuable advancement for computational biomolecular simulations.