Membrane proteins are essential to many cellular activities and their defects may cause disease. A remaining challenge in homology modeling of helical membrane proteins is to improve the accuracy of models built based on templates of the low sequence identity. The long-term objective of this project is to develop novel approaches to address this challenge. For this proposal, proposed works will: 1) Optimize and verify the scoring function based on inter-residue interactions for evaluating computationally built membrane protein models. In our previous studies, we've developed a scoring function based on inter-residue interactions for the model evaluation. Here we aim to optimize and verify this function so that it is of best use in the following studies. 2) Characterize conserved interactions among homologous protein structures through structure analysis and sequence-based conservation analysis. In this part, we propose to: (i) Confirm the generality of the existence of a set of conserved inter-residue interactions for any protein family;(ii) Demonstrate that the conserved interactions across homologous protein structures can be identified from a multiple sequence alignment of homologous proteins with >35% sequence identity, as that alignment can be obtained with high accuracy;(iii) Elucidate conserved interactions for 28 pairs of homologous transmembrane (TM) protein structures with low sequence identity that will be used in the following studies. 3) Develop and validate a novel approach for improving homology modeling for helical membrane proteins that complements and is built upon the available software. Built on the work from Specific Aim I and II, and taken the 28 pairs of homologous TM proteins as examples, we'll test the approach to refine the sequence alignment in homology modeling by incorporating conserved interactions into the modeling process. For each pair, one will be adopted as the template to construct the model for the other using the available state-of-art homology-modeling software. Each refined alignment will be compared with the best alignment obtained by superposition of the X-ray structures of the template and the target proteins to validate the proposed modeling approach. The proposed work represents a novel approach for improving homology modeling techniques in order to broaden its application for helical membrane proteins that can be implemented int modeling approaches in place.
Membrane proteins account for approximately 30% of the human genome and are targets of roughly 50% of drugs currently available in the market. However, there are only about 100 high-resolution structures determined experimentally for them, mainly due to technical difficulties. Homology modeling is the only technique that can generate high-quality models comparative to experimentally determined structures. Improving this technique to expand its application for modeling of helical membra ne proteins based on templates of low sequence identity is thus critical to the studies of biological roles of membrane proteins and to the development of new drugs targeting them.
| Chaudhari, Rajan; Heim, Andrew J; Li, Zhijun (2015) Improving homology modeling of G-protein coupled receptors through multiple-template derived conserved inter-residue interactions. J Comput Aided Mol Des 29:413-20 |
| Heim, Andrew J; Li, Zhijun (2012) Developing a high-quality scoring function for membrane protein structures based on specific inter-residue interactions. J Comput Aided Mol Des 26:301-9 |
| Pabuwal, Vagmita; Li, Zhijun (2011) Comparison analysis of primary ligand-binding sites in seven-helix membrane proteins. Biopolymers 95:31-8 |
