Exploiting Serial Crystallography in Structure-based Drug Design

Abstract

Studying the interaction between ligands and proteins is an important step in the Structure-Based Drug Design (SBDD) process. The most widely used method for these studies is X-ray crystallography. However, depending on the specific characteristics of the system being studied, some ligands may not fully penetrate into the large macromolecular crystals that are typically used for these studies. Microcrystals offer the possibility of more complete soaking of ligands, but have several technical challenges. The first is that until recently, access to instrumentation for routine microcrystallography has been limited. Such experiments require highly focused and intense X-ray beams that match the size of protein crystals, as well as extremely high precision goniometry to ensure that the sample stays centered in the beam. Additionally, radiation damage generally prevents the collection of full datasets from individual crystals. This means that a serial approach must be employed in which data from multiple crystals are combined. In many cases this is a trivial task, but in some cases, these microcrystals can exhibit non-isomorphism, possibly due to natural variations or binding of heavy atoms or compound(s). We used techniques to deal with non isomorphism such as a genetic algorithm (CODGAS) or hierarchical cluster analysis (HCA) to segregate serial crystallographic data in order to improve the electron density of bound ligands. Protocols were developed and tested both on experimental and simulated data. In both cases, we observed improvements in the ligand's electron density amongst the separated groups, compared to merging all sub datasets. This improvement could facilitate the modeling of protein-ligand interactions, which could in turn facilitate the drug discovery process.

Zoom link

Please click the link below to join the webinar:
https://esrf.zoom.us/j/97283711724?pwd=TVhiMkpJNUlXSFhyTkpxdWUwYlhhZz09
Passcode: 894526