Quantum Computational Quantification of Protein-Ligand Interactions

A prototypical hybrid classical and quantum computational tools stack for the quantification of protein-ligand interactions is presented herein by using binding energy differences to rank-order a series of β-secretase (BACE1) inhibitors. Due to the large Hilbert space involved, the combination of QM/MM and Density Matrix Embedding Theory (DMET) procedure involving reduced active orbital space with the Variational Quantum Eigensolver (VQE) approach is proven to be effective for finding the ground states of such complex systems. To reduce the device-induced noise, two different techniques were employed: (i) the State Preparation and Measurement (SPAM) error mitigation and (ii) a new flavour of Symmetry Verification approach, the in-house developed Partition Measurement Symmetry Verification (PMSV) algorithm, which is unique in the sense that it does not require any extra quantum resources. The experiments were conducted on the latest superconducting transmon (IBM) and trapped-ion (Honeywell) Noisy Intermediate Scale Quantum (NISQ) devices. This is the first application of real quantum computers to the calculation of protein-ligand interaction. The results shed light on hardware and software requirements which would enable the application of NISQ algorithms in drug design.