I am happy to announce a new publication available in Chemical Science entitled Discovery of SARS-CoV-2 Mpro peptide inhibitors from modelling substrate and ligand binding. This was a collaborative effort involving many people. Our study involved both experimental and theoretical work (docking, classical molecular dynamics, QM/MM, linear-scaling DFT, etc). My contribution, of course, was on the linear-scaling DFT side of things.
The main protease of SARS-CoV-2 (Mpro) plays a central role in replication of the virus. Mpro works as a kind of scissors that cuts up large polypeptides in specific places. If one can design a drug that binds strongly to the protease without being degraded, the protease can be deactivated, which would prevent the virus from replicating. In this paper, we studied the binding mechanism of the protease in order to rationally design potential drugs. We examined both how it binds to the cleavage sites of the polypeptides and a large database of small molecule inhibitors. This analysis gave us design rules that helped us design new peptides which effectively inhibit the protease.
I am curious what the reader experience will be like with this paper. I imagine that some readers will focus in on their particular expertise, and find themselves wishing we presented more calculations. For a multi-disciplinary paper like this though, I think quantity is its own kind of quality. You have to find the signal that is coming through from the sum total of your analysis techniques. This isn't to say each step wasn't caried out thoroughly, but just a reminder that each technique is merely a tool which helps you get to that final step of designing and experimentally testing the new drug. The paper is available as open access so I hope you give it a read.