Karplus K, Hu B
Computer Engineering, University of California, Santa Cruz, 59064, USA.
Bioinformatics. 2001 Aug;17(8):713-20. doi: 10.1093/bioinformatics/17.8.713.
SAM-T99 is an iterative hidden Markov model-based method for finding proteins similar to a single target sequence and aligning them. One of its main uses is to produce multiple alignments of homologs of the target sequence. Previous tests of SAM-T99 and its predecessors have concentrated on the quality of the searches performed, not on the quality of the multiple alignment. In this paper we report on tests of multiple alignment quality, comparing SAM-T99 to the standard multiple aligner, CLUSTALW.
The paper evaluates the multiple-alignment aspect of the SAM-T99 protocol, using the BAliBASE benchmark alignment database. On these benchmarks, SAM-T99 is comparable in accuracy with ClustalW.
The SAM-T99 protocol can be run on the web at http://www.cse.ucsc.edu/research/compbio/HMM-apps/T99-query.html and the alignment tune-up option described here can be run at http://www.cse.ucsc.edu/research/compbio/HMM-apps/T99-tuneup.html. The protocol is also part of the standard SAM suite of tools. http://www.cse.ucsc.edu/research/compbio/sam/
SAM-T99是一种基于迭代隐马尔可夫模型的方法，用于寻找与单个目标序列相似的蛋白质并进行比对。其主要用途之一是生成目标序列同源物的多序列比对。之前对SAM-T99及其前身的测试集中在搜索的质量上，而非多序列比对的质量。在本文中，我们报告了多序列比对质量的测试，将SAM-T99与标准多序列比对工具CLUSTALW进行比较。
本文使用BAliBASE基准比对数据库评估了SAM-T99协议的多序列比对方面。在这些基准测试中，SAM-T99在准确性上与ClustalW相当。
