Model Quality Assessment for Membrane Proteins Based on Nonlinear Dimension Reduction Methods
This post-doc project is in collaboration with the Theoretical and Computational Biophysics group at KTH.
Zhijie Wen is the post-doc. The project involves Erik Lindahl, Arjun Ray and Björn Wallner at Theoretical and Computational Biophysics group at KTH, and Jan-Olov Strömberg (CIAM/KTH) and Ozan Öktem (CIAM/KTH).
The project is funded through CIAM.
Learning-based model quality assessment programs have been quite successful at discriminating between high- and low-quality protein structures. The performance can be significantly improved by restricting the learning space to a specific context, in this case membrane proteins. Since these are among the most important structures from a pharmaceutical point-of-view, it is particularly interesting to resolve local model quality for regions corresponding, e.g. to binding sites. In a publication , the Theoretical and Computational Biophysics group use a support vector machine with a combination of general and membrane protein-specific features.
The project seeks to explore the usage of non-linear dimension reduction methods as an alternative to support vector machine based methods. The initial idea of using model quality assessment is computationally much more demanding and seems not to provide enough benefits when compared against support vector machine based methods. A paper summarised the results is in preparation.