Program in Biomedical Informatics, Stanford University, Stanford, CA, 94305 USA.
Genome Biol. 2008 Jan 16;9(1):R8. doi: 10.1186/gb-2008-9-1-r8.
Structural genomics efforts have led to increasing numbers of novel, uncharacterized protein structures with low sequence identity to known proteins, resulting in a growing need for structure-based function recognition tools. Our method, SeqFEATURE, robustly models protein functions described by sequence motifs using a structural representation. We built a library of models that shows good performance compared to other methods. In particular, SeqFEATURE demonstrates significant improvement over other methods when sequence and structural similarity are low.
结构基因组学研究导致了越来越多新颖的、未被描述的蛋白质结构，它们与已知蛋白质的序列同一性较低，因此需要越来越多基于结构的功能识别工具。我们的方法 SeqFEATURE 使用结构表示形式稳健地对序列基序描述的蛋白质功能进行建模。我们构建了一个模型库，与其他方法相比，该模型库表现出了良好的性能。特别是，当序列和结构相似性较低时，SeqFEATURE 比其他方法有显著的改进。
