New method uses quantum and classical tools to model ligand shapes

A new QUBO-based approach selects ligand conformations using a quantum-classical hybrid solver.

A ligand can adopt multiple conformations within the same binding site. These conformations can interact differently with the surrounding protein.

Multiconformer modeling must determine which candidate conformations to retain and what occupancies to assign to them. This study reformulated conformer selection as a quadratic unconstrained binary optimization (QUBO) problem.

The researchers used a quantum-classical hybrid solver to solve the QUBO. They compared this workflow with the mixed-integer quadratic programming (MIQP) based qFit-ligand workflow.

The comparison used 83 paired X-ray ligand entries. Quantum-qFit produced higher ligand-region real-space correlation coefficients (RSCC) in 62 entries, which is 74.7%.

Quantum-qFit also showed lower occupancy-weighted torsional strain in 66 entries, or 79.5%. The molecular Electron Density for Individual Atoms (EDIAm) score decreased in most entries.

The results show that the conformer-selection formulation can change which ligand states are represented from the same experimental data. This provides a QUBO framework for studying multiconformer ligand selection.

Key facts

  • Quantum-qFit produced higher RSCC in 74.7% of the 62 entries.
  • The workflow showed lower occupancy-weighted torsional strain in 79.5% of 66 entries.
  • The EDIAm score decreased in most entries.

All sections

Edition