Construction of Database of Rat and Mouse Genes for Three Dimensional Monomer and Complex Protein Structure

首先，我们改进了我们的全自动建模系统（FAMS）和蛋白质复合物全自动建模系统（FAMS Complex）的性能。这些建模系统使得预测精确的三维蛋白质结构成为可能。FAMS在第七届蛋白质结构预测技术关键评估社区实验（CASP 7）中表现出成功的蛋白质三维结构预测性能，该实验评估了建模方法的准确性。此外，FAMS Complex在蛋白质-蛋白质对接结构预测（卡普里）的社区范围比较评价实验中表现良好。其次，我们开发了蛋白质结构预测和氨基酸序列配对比对的评价系统。最后，构建了大鼠和小鼠基因的蛋白质结构建模数据库，并以RIKEN FAMSBASE（http：//famshelp.gsc.riken.jp/famsbase/）的形式对外发布。在RIKEN FAMSBASE中，任何研究人员都可以从氨基酸序列，基因名称和关键字搜索大鼠和小鼠基因的三维蛋白质结构模型。

In the first, we improved the performance of our Full Automatic Modeling System (FAMS) and Full Automatic Modeling System for protein complex (FAMS Complex). These modeling systems made it possible to predict accurate three-dimensional protein structures. The FAMS showed successful performance for prediction of protein three-dimensional structures in 7th Community Wide Experiment on the Critical Assessment of Techniques for Protein Structure Prediction (CASP7), which assessed the accuracy of modeling methods. Moreover the FAMS Complex performed well in Community wide experiment on the comparative evaluation of protein-protein docking for structure prediction (CAPRI).Secondarily, we developed the evaluation systems which evaluate quality of predicted protein structures and pairwise-alignments of amino acid sequences. These methods provide information of confidence of predicted models.Finally, we constructed huge protein structure modeling database of Rat and Mouse genes, and released as RIKEN FAMSBASE (http://famshelp.gsc.riken.jp/famsbase/) to the public. In the RIKEN FAMSBASE, any researchers can search three-dimensional protein structure models of Rat and Mouse genes from amino acid sequence, a Gene Name, and keywords.