Proteins from yeast are chosen to represent evolutionarily diverse families whose members are not predicted to correspond to a known protein structure (based on Sali Lab Yeast Protein Models). The initial protein families were selected from clusters generated on the basis of sequence similarity by SYSTERS, PRODOM, and DOMO. Coding sequences obtained by PCR from genomic DNA are cloned, expressed, purified and crystallized for structure determination. Information about the proteins is provided and progress toward determination of the structures is summarized in the data tables. Proteins P013-P016 were chosen to test whether the structure prediction tools did in fact discover structural similarity between proteins that align with only 14-22% identity in amino-acid sequence. The primary objective is to put in place the technology needed to increase knowledge of the diversity of protein folds in nature. This basic information is needed to develop computational tools for predicting protein structures from gene sequences and for predicting protein function from structure. These tools will ultimately be applied to understand the functions and interactions of the 100,000 or so proteins of the human proteome. Information on the project is available at http://genome5.bio.bnl.gov/Proteome/
Showing the most recent 10 out of 68 publications