QUBO Problem Formulation of Fragment-Based Protein-Ligand Flexible Docking.

Area of Science:

• Computational chemistry
• Structural biology
• Quantum computing

Background:

• Protein-ligand docking is crucial for structure-based drug discovery, estimating binding modes and energies.
• Previous quantum annealing approaches for docking overlooked essential internal compound flexibility.

Purpose of the Study:

• To formulate a fragment-based protein-ligand flexible docking method using a QUBO problem.
• To incorporate internal degrees of freedom into quantum-annealed docking calculations.

Main Methods:

• Developed a QUBO formulation for flexible docking, focusing on rigid chemical fragments.
• Incorporated interaction energy, fragment clashes, covalent bonds, and placement constraints into the Hamiltonian.
• Utilized a simulated quantum annealer (SQBM+) for proof-of-concept redocking of Aldose reductase.

Main Results:

• The QUBO formulation successfully predicted a near-native binding pose (RMSD = 1.26 Å) for the Aldose reductase-compound complex.
• Post-docking energy minimization further refined the pose to high accuracy (RMSD = 0.27 Å).

Conclusions:

• The proposed QUBO formulation for fragment-based flexible docking is valid and effective.
• This approach advances quantum annealing applications in drug discovery by handling molecular flexibility.