Turning high-throughput structural biology into predictive inhibitor design

DOI: 10.1073/pnas.2214168120

Authors: William Mccorkindale (University of Cambridge) , Kadi L. Saar (University of Cambridge) , Daren Fearon (Diamond Light Source) , Melissa Boby (Sloan Kettering Institute) , Haim Barr (The Weizmann Institute of Science) , Amir Ben-Shmuel (Israel Institution of Biological Research) , Nir London (Weizmann Institute of Science) , Frank Von Delft (Diamond Light Source; University of Oxford; University of Johannesburg; Research Complex at Harwell) , John D. Chodera (Sloan Kettering Institute) , Alpha. A. Lee (University of Cambridge) , The Covid Moonshot Consortium
Co-authored by industrial partner: No

Type: Journal Paper
Journal: Proceedings Of The National Academy Of Sciences , VOL 120

State: Published (Approved)
Published: March 2023

Abstract: A common challenge in drug design pertains to finding chemical modifications to a ligand that increases its affinity to the target protein. An underutilized advance is the increase in structural biology throughput, which has progressed from an artisanal endeavor to a monthly throughput of hundreds of different ligands against a protein in modern synchrotrons. However, the missing piece is a framework that turns high-throughput crystallography data into predictive models for ligand design. Here, we designed a simple machine learning approach that predicts protein–ligand affinity from experimental structures of diverse ligands against a single protein paired with biochemical measurements. Our key insight is using physics-based energy descriptors to represent protein–ligand complexes and a learning-to-rank approach that infers the relevant differences between binding modes. We ran a high-throughput crystallography campaign against the SARS-CoV-2 main protease (MPro), obtaining parallel measurements of over 200 protein–ligand complexes and their binding activities. This allows us to design one-step library syntheses which improved the potency of two distinct micromolar hits by over 10-fold, arriving at a noncovalent and nonpeptidomimetic inhibitor with 120 nM antiviral efficacy. Crucially, our approach successfully extends ligands to unexplored regions of the binding pocket, executing large and fruitful moves in chemical space with simple chemistry.

Journal Keywords: machine learning; drug design; crystallography

Diamond Keywords: COVID-19; Viruses

Subject Areas: Biology and Bio-materials, Medicine, Information and Communication Technology

Diamond Offline Facilities: XChem
Instruments:

Added On: 13/03/2023 09:02

Documents:
pnas.2214168120.pdf

Discipline Tags:

Pathogens Infectious Diseases Artificial Intelligence Health & Wellbeing Information & Communication Technologies Structural biology Data processing Drug Discovery Life Sciences & Biotech

Technical Tags:

Diffraction Macromolecular Crystallography (MX) Fragment Screening