Development of Methods to Combine Sequence, Structure, and Network Analyses to Study Gene Annotation, Coevolution, and Allostery in Protein:RNA Complexes
Title: Development of Methods to Combine Sequence, Structure, and Network Analyses to Study Gene Annotation, Coevolution, and Allostery in Protein:RNA Complexes.
Speaker: Dr. Anurag Sethi, Theoretical Biology and Biophysics Group, Theoretical Division, Los Alamos National Laboratory
Abstract: Combination of sequence and structure information to generate statistically unbiased sets of data to gain insight about the evolution of a biomolecular family is an important issue in bioinformatics. We have developed an algorithm, which is based on the multidimensional QR factorization of sequence and structure alignments, to generate complete evolutionary profiles that represent the topology of the molecular phylogenetic tree of the homologous group. Merging the structure and sequence information allows the construction of accurate profiles for distantly related groups. These structure-based profiles outperform other sequence-based methods for finding distant homologs and were used to identify an archaeal aminoacyl-tRNA synthetase (aaRS) that is involved in cysteine aminoacylation and had eluded previous annotation studies [PNAS, 4045 (2005) & PNAS , 19003 (2005)]. This algorithm has also been used to show that conserved residues form the most energetic contacts in a complex of elongation factor-Tu with the charged tRNA [JMB, 1382, (2008)] and to elucidate that regions in the ribosomal proteins and ribosomal RNA have coevolved with each other in the different domains of life [PNAS, 13953 (2008)]. Finally, weighted networks based on the dynamic contact map of a protein:RNA complex during a molecular dynamics simulation were used to identify the optimal and suboptimal paths for communication across regions in the complex that are 40-70 Angstroms apart. The communication between monomers in different communities forms the bottleneck for signal transmission within the network and were used to identify residues and nucleotides that are predicted to be critical for allostery in the complex [PNAS, 6620 (2009)].
