Interaction Energy Decomposition

Rowan's interaction-energy-decomposition workflow calculates the interaction of one fragment with its surrounding system and decomposes that interaction energy into interpretable components like electrostatic, exchange, dispersion, and induction interaction energies.

Background

While it's relatively simple to calculate the total interaction energy of a dimeric system (like a protein–ligand complex or a small-molecule dimer) with quantum-mechanical (QM) methods, the results of basic QM-style calculations don't neatly map onto the concepts chemists think about when it comes to interactions like van der Waals forces, π–π interactions, and Coulombic forces.

Interaction-energy-decomposition calculations help span this gap by splitting a QM energy result into conceptual buckets so that interactions can be reasoned about analytically. In drug discovery contexts, this can be used to rationally guide the optimization of a ligand's on-target potency in a structurally enabled campaign.

How it works

Rowan currently supports the SAPT0 method (implemented in Psi4) for interaction-energy-decomposition calculations.

Symmetry-adapted pertubation theory (SAPT) is an approach to the general energy decomposition analysis problem; it calculates interaction energies via perturbation, causing nice cancelation of errors. SAPT0 is a fast SAPT method that can be applied to large systems as well as open-shell systems. SAPT0 calculations employ the jun-cc-pVDZ basis set (a truncated version of aug-cc-pVDZ) as recommended by Parker et al. 2014.

SAPT0 calculations return an overall interaction energy between a specified fragment and the rest of the input system, and they decompose the interaction energy into the following terms:

  • Electrostatic: attraction/repulsion between partial charges on each system.
  • Exchange: Pauli exchange due to antisymmetrization (a purely quantum effect).
  • Dispersion: electron correlation.
  • Induction: polarization of each monomer by the electric field of the other (includes exchange-induction).

A more thorough introduction to SAPT methods and SAPT0 specifically is available in the first seven minutes of this video by Psi4 contributor Zach Glick.