Multiple Simultaneous Alignment and Ab Initio Structure Prediction of Pseudoknotted RNAs and Proteins

• Applicant: Dr. Sebastian Will
• Subject Area: Bioinformatics and Theoretical Biology
• Term: from 2009 to 2011
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 157432568

Structure analysis of RNA and proteins is a central task in biological research. However, this analysis is difficult for low-homology RNA or protein classes. Proteins as well as RNAs with pseudoknots (PK RNA) show crossing interactions, which makes their analysis algorithmically challenging. I propose to develop ab initio solutions to the central problems in the analysis of PK RNAs and transmembrane proteins (TMs), which form a biomedically important protein class. I will consider the problems of structure prediction and multiple structural alignment. The methods will exploit specific restrictions of the structure, as they are found for most PK RNAs and TMs that show a regular interaction pattern. This allows to transfer and extend techniques from (pseudoknot-free) RNA analysis, which offers many well-established, efficient methods. I will solve both problems simultaneously by predicting a comparatively supported common structure for a set of related, unaligned TM proteins or RNAs. Therefore, I will develop a multiple alignment procedure based on pairwise simultaneous alignment and folding. The structure-enhanced multiple alignment is finally used for structure inference. Thus, the project will provide new algorithmic support and practicable tools for the structural analysis of TMs and PK RNAs. Studying proteins and PK RNA will further our understanding of crossing structure in general.

• DFG Programme: Research Fellowships
• International Connection: USA
• Host: Professorin Bonnie Berger