Bioinformatics linkage of protein disorder and function

点击翻译按钮获取中文摘要

The overall goal of this project is to better understand the relationship between intrinsic disorder (ID) and protein function. Much has been learned since the the first submission of the proposal that became the grant for the first 4 years of this project. This is due to the activity of our research team supported by the previous grant and also due to the significantly intensified research of other laboratories on the intrinsically disordered proteins (IDPs). However, many important questions remained unanswered as of yet. Therefore, this proposal is focused on five major problems: 1) Association between protein ID and alternative splicing (AS) of pre-mRNA (AS is believed to represent a crucial event responsible for the functional profiling of many protein in eukaryotes, including proteins in such important pathways as Notch, Wnt, Hedgehog, PAR, and TGF ); 2) Characterization of disorder-to-order transitions induced in IDPs by interaction with their binding partners (with the major focus on the development of predictors of molecular recognition features (MoRFs) from sequence); 3) Understanding sequence-function relationships for ID regions associated with specific functions (with the primary focus on the discovery of novel functions associated with intrinsic disorder and finding functions specific for MoRFs, ELMs, Fmotifs, SMART, or Pfam motifs); 4) Characterization of the relationships between ID and single nucleotide polymorphisms (SNPs) that occur in coding regions (to check a hypothesis that a large fraction of disease related SNPs are located in ordered regions, since a substitution of a buried amino-acid is more likely to disrupt the structure stability of a protein and alter its function); and 5) Characterization of protein-RNA interactions (to test the hypothesis that the protein-RNA interactions frequently involve disorder-to-order transitions of the protein components). For each of these aims, a set of specific proteins will be collected or significantly enlarged with the major emphasis on the exhaustive and thorough annotation of disordered and ordered regions. Using these annotated data, we propose next to compare different bioinformatics and datamining strategies to find the optimal approaches for the disorder/order problem. The proposed research has important implications for human diseases, since many proteins involved in numerous human diseases (including cancer and cardiovascular diseases) have significant ID regions and their structures and functions of are modulated by AS and SNPs.