Samudrala R, Moult J
Center for Advanced Research in Biotechnology, University of Maryland Biotechnology Institute, Rockville 20850, USA.
Proteins. 1997;Suppl 1:43-9. doi: 10.1002/(sici)1097-0134(1997)1+<43::aid-prot7>3.3.co;2-z.
We constructed five comparative models in a blind manner for the second meeting on the Critical Assessment of protein Structure Prediction methods (CASP2). The method used is based on a novel graph-theoretic clique-finding approach, and attempts to address the problem of interconnected structural changes in the comparative modeling of protein structures. We discuss briefly how the method is used for protein structure prediction, and detail how it performs in the blind tests. We find that compared to CASP1, significant improvements in building insertions and deletions and sidechain conformations have been achieved.
在蛋白质结构预测方法关键评估（CASP2）的第二次会议上，我们以盲法构建了五个比较模型。所使用的方法基于一种新颖的图论团簇查找方法，试图解决蛋白质结构比较建模中相互关联的结构变化问题。我们简要讨论了该方法如何用于蛋白质结构预测，并详细说明了它在盲测中的表现。我们发现，与CASP1相比，在构建插入、缺失和侧链构象方面取得了显著改进。
