ADVANCED SEQUENCE BASED PREDICTION OF PROTEIN FUNCTION

描述（改编自申请人摘要）： 该项目的目标是自动化和提高预测 基于多序列数据的蛋白质结构和功能。 具体 目的是：1）将整个序列数据库划分为一个综合的 一组同源蛋白质结构域; 2）设计先进的，家族特异性的 这些域的多个比对模型; 3）开发统计 序列与模型相似性显著性的估计方法 评分; 4）设计相应的比对优化程序;以及5） 开发基于蛋白质结构和功能方面的预测工具 在这些路线上。 方法包括吉布斯抽样和隐马尔可夫模型 多序列比对的方法，结构穿线方法， 动态编程和类似BLAST的数据库搜索程序等 统计和算法方法。 蛋白质的综合数据库 生物医学界将通过 国家生物技术信息中心。

DESCRIPTION (Adapted from the applicant's abstract): The goal of this project is to automate and improve the prediction of protein structure and function based on multiple sequence data. Specific aims are to: 1) partition the entire sequence database into a comprehensive set of homologous protein domains; 2) design advanced, family-specific multiple alignment models of these domains; 3) develop statistical procedures for estimating the significance of sequence-to-model similarity scores; 4) devise corresponding alignment optimization procedures; and 5) develop tools for predicting aspects of protein structure and function based on these alignments. Methods include Gibbs sampling and hidden Markov model procedures for multiple sequence alignment, structural threading methods, dynamic programming and BLAST-like database search procedures, and other statistical and algorithmic methods. A comprehensive database of protein domains will be made available to the biomedical community through the National Center for Biotechnology Information.