基于矩阵打分值和化学位移值预测酶蛋白质中β-发夹模体  

Prediction of β-hairpin Motif in Enzyme Based on Values of Matrix Scoring and Averaged Chemical Shifts

作　　者：宋航宇 胡秀珍[1] 

机构地区：[1]内蒙古工业大学理学院,呼和浩特010051

出　　处：《内蒙古工业大学学报（自然科学版）》2014年第3期174-181,共8页Journal of Inner Mongolia University of Technology：Natural Science Edition

基　　金：国家自然科学基金(30960090;31260203)

摘　　要：酶是一类具有催化功能的蛋白质,对其结构和功能的研究是非常有必要的。特殊模体β-发夹是一种简单的超二级结构,它们包含了丰富的折叠信息,对酶中β-发夹的预测有助于酶结构和功能的预测。本文建立了非冗余的β-发夹模体数据集,包含1080条酶蛋白质链,2818个β-发夹模体,1098个非β-发夹。提取位点亲疏水组分及位点亲疏水紧邻关联组分作为参数,运用矩阵打分算法得到的预测结果不理想。将位点亲疏水、位点亲疏水紧邻关联组分为参数得到的打分值和平均化学位移值共同作为特征参数,输入支持向量机算法对模体进行预测,5交叉检验预测总精度是81.8%,相关系数达到了0.636。Enzyme is a kind of protein that acts with catalytic functions in living organisms,thus the study of the structure and function of enzyme is essential.β-hairpin is a specific super-secondary structure motif which carries abundant folding information.The prediction ofβ-hairpin motifs in enzyme protein is helpful to study the structure and function of enzyme.In this paper,we constructed a non-redundantβ-hairpin database,which have 2,818β-hairpins and 1,098non-β-hairpins motif in 1,080 enzyme protein chains.The hydropaths composed of position and adjacent to dipeptide hydropaths composition of position extracted as parameter for matrix scoring algorithm,the prediction results were unsatisfactory.Then,we selected the scores of hydropathy composition of position,the scores of adjacent to dipeptide hydropathy composition of position and averaged chemical shift as the input parameters for Support Vector Machine algorithm,the overall accuracy and Matthew＇s correlation coefficient reached 81.8%and 0.636 respectively by 5-fold cross-validation.

关 键 词：Β-发夹模体 支持向量机算法 矩阵打分算法 位点亲疏水组分 位点亲疏水关联 平均化学位移值 

分 类 号：Q61[生物学—生物物理学]