Short Report by Axel Fischer & Gregory Sliwoski: INTEGRATION OF LIMITED EXPERIMENTAL DATA INTO ENSEMBLE PREDICTION at Leipzig University (Nov 14 – Dec 22)

/ February 15, 2016/ News, Students

The objective of our visit was deepening the collaboration between Leipzig University and Vanderbilt University regarding the integration of limited experimental data into protein ensemble prediction.

During this visit Axel Fischer presented a method to integrate cross-linking data into protein structure prediction methods at the 5th Symposium on Structural Proteomics in Halle. Following subsequent discussions with Dr. Stefan Kalkhof and Tommy Hofmann, work on an algorithm to efficiently simulate cross-link conformations was continued. In a preliminary study, the algorithm significantly improves the accuracy with which protein structures can be predicted.

Additionally, Axel developed an algorithm to integrate solid-state nuclear magnetic resonance (ssNMR) data into the protein structure prediction software BCL::Fold. While the algorithm still has to thoroughly benchmarked on experimental data, incorporation of ssNMR data will provide information about the tilt angles and rotation states of transmembrane helices, which will improve the accuracy of protein structure prediction methods.

Gregory Sliwoski met with collaborators Prof. Dr. Torsten Schöneberg and Dr. Antje Brüser to discuss the experimental results of mutations suggested by computational ligand-receptor models and devised strategies to incorporate these results into further rounds of protein modeling and ligand docking. Additionally, Gregory successfully defended his thesis titled “Computer Aided Drug Discovery: Descriptor improvement and application to obesity-related therapeutics” at the Universität Leipzig.

During this stay we enjoyed Gluehwein at the Christmas Market at the Connewitzer Kreuz. We also visited the Voelkerschlacht Denkmal. Managing to climb to its top provides a beautiful view over Leipzig.