Lab of Computational Chemistry and Drug Design, Laboratory of Chemical Genomics, Peking University Shenzhen Graduate School, Shenzhen, PR China.
PLoS One. 2013 Jun 11;8(6):e65705. doi: 10.1371/journal.pone.0065705. Print 2013.
WD40-repeat proteins (WD40s), as one of the largest protein families in eukaryotes, play vital roles in assembling protein-protein/DNA/RNA complexes. WD40s fold into similar β-propeller structures despite diversified sequences. A program WDSP (WD40 repeat protein Structure Predictor) has been developed to accurately identify WD40 repeats and predict their secondary structures. The method is designed specifically for WD40 proteins by incorporating both local residue information and non-local family-specific structural features. It overcomes the problem of highly diversified protein sequences and variable loops. In addition, WDSP achieves a better prediction in identifying multiple WD40-domain proteins by taking the global combination of repeats into consideration. In secondary structure prediction, the average Q3 accuracy of WDSP in jack-knife test reaches 93.7%. A disease related protein LRRK2 was used as a representive example to demonstrate the structure prediction.
WD40 重复蛋白(WD40s)作为真核生物中最大的蛋白质家族之一,在组装蛋白-蛋白/DNA/RNA 复合物中起着至关重要的作用。WD40s 尽管序列多样化,但折叠成相似的β-桨叶结构。已经开发了一个 WDSP(WD40 重复蛋白结构预测器)程序,用于准确识别 WD40 重复并预测它们的二级结构。该方法通过结合局部残基信息和非局部家族特异性结构特征,专门为 WD40 蛋白设计。它克服了高度多样化的蛋白质序列和可变环的问题。此外,WDSP 通过考虑重复的全局组合,在识别多个 WD40 结构域蛋白方面实现了更好的预测。在二级结构预测中,WDSP 在自举测试中的平均 Q3 准确率达到 93.7%。以疾病相关蛋白 LRRK2 为例,说明了结构预测。