by: RANDRIHARIMANAMIZARA Ulrick Fineddie Track « In Silico Drug Design », Strasbourg-Milan-Paris, 2022
In the present interview, we will talk about Docking, its success and challenges with an expert and some advices for students and future scientists who are interested in this subject.
1) Could you please introduce yourself and tell a little bit about your scientific career?
My name is Peter Kolb and I am a professor of Pharmaceutical Chemistry in Marburg. I started out as a biochemist, but early on developed a great interest in theoretical chemistry, chemoinformatics, etc. It simply seemed more reproducible to be able to control all variables. I still did my diploma thesis on an experimental subject, but then switched to computational methods for my doctoral thesis. However, I always did at least part of the experimental verification of a project myself and I still think that it is important to always verify calculations experimentally.
2) Could you explain in a few words what “docking” is and tell us one success story of it that has impressed you?
Docking is a computational method that tries to find a plausible pose for a small-molecule ligand in a protein binding site and numerically evaluates the goodness of fit. Thus, entire libraries of candidate ligands can be ranked (high throughput mode) or individual ligands computationally modified and optimized. I think the biggest successes are where docking finds ligands that look completely different than already known ones. See refs 1-3 for examples.
3) In your opinion, what is the main challenge in docking so far?
The many false positives. In order for docking to be as fast as it is, scoring functions have to be very simple. This also makes them imprecise. Thus, when one docks large libraries, many of the top-ranked molecules will actually not bind to the protein of interest in an experiment.
4) Can you say a few words about one of your research areas now?
We are currently most interested in G protein-coupled receptors, membrane-bound signalling proteins. These are highly allosteric proteins and we are trying to understand how their activity might be influenced through allosteric sites. Once we have identified such a site, we try to find out whether it can host ligands, what natural ligands might be, and how these ligands change the activity of the receptor.
5) What advice can you give for students who want to get into this domain?
A solid background in chemistry is key, as well as good knowledge in pharmacology. One needs less programming skills than one might assume :) When choosing a research topic, I suggest to focus on a biological (sub)question beyond a simple screen for ligands.
 Levit Kaplan, A., Confair, D.N., Kim, K. et al. Bespoke library docking for 5-HT2A receptor agonists with antidepressant activity. Nature 610, 582-591 (2022). https://doi.org/10.1038/s41586-022-05258-z
 Lyu, J., Wang, S., Balius T.E. et al. Ultra-large library docking for discovering new chemotypes. Nature 566, 224–229 (2019). https://doi.org/10.1038/s41586-019-0917-9
 Kolb, P., Rosenbaumb, D.M., Irwin, J.J. et al. Structure-based discovery of beta2-adrenergic receptor ligands. PNAS 106, 6843– 6848 (2009). https://doi.org/10.1073/pnas.081265710