接触图辅助的过程重采样蛋白质构象空间优化算法

Contact Map-assisted Process Resampling Protein Conformation Space Optimization Algorithm

蛋白质的三维结构是研究其生物功能及活性机理的基础.为了提高蛋白质结构的预测精度,在进化计算的框架下,提出一种接触图辅助的过程重采样蛋白质构象空间优化算法(Contact Map-assistedProcess Resampling Protein Conformation Space Optimization Algorithm,CM PR). CM PR算法基于残基接触图设计打分模型,用于选择构象以构建过程片段库,使用基于过程重采样策略的片段组装技术执行变异操作,残基接触先验知识和种群进化过程统计知识辅助采样,可以增强近天然态构象区域的搜索能力,提高蛋白质结构预测精度.在12个测试蛋白上的实验结果表明,所提方法具有良好的近天然态构象采样能力和较高的预测精度.

Protein 3 D structure is the basis for studying its biological function and activity mechanism. In order to improve the prediction accuracy of protein structure,a contact map-assisted process resampling protein conformation space optimization algorithm( CMPR)was proposed in the framework of evolutionary computation. The residue contact map based scoring model in CMPR is designed to select the conformation for constructing the process fragment library,the process resampling strategy based fragment assembly technique is used for conducting the mutation operation,the residue contact prior knowledge and the population evolution process statistical knowledge are used for assisting the sampling process,the ability for sampling the near-native conformation regions can be enhanced,thus the accuracy of protein structure prediction can be improved. The experimental results on 12 test proteins show that the proposed method has good near-native conformation sampling ability and high prediction accuracy.