Tong Jian-Bo, Zhang Sheng-Wan, Cheng Su-Li, Li Gai-Xian
College of Chemistry and Chemical Engineering, Shanxi University, Taiyuan 030006, China.
Yao Xue Xue Bao. 2007 Jan;42(1):40-6.
To establish a new amino acid structure descriptor that can be applied to polypeptide quantitative structure activity relationship (QSAR) studies, a new descriptor, SVRDF, was derived from a principal components analysis of a matrix of 150 radial distribution function index of amino acids. The scale was then applied in three panels of peptide QSAR that were molded by partial least squares regression. The obtained models with the correlation coefficients (R2(cum)), cross-validation correlation coefficients (Q2(cum)) were 0.766 and 0.724 for 48 bitter tasting dipeptides; 0.941 and 0.811 for 21 oxytocin analogues; 0.996 and 0.919 for 20 thromboplastin inhibitors. Satisfactory results showed that information related to biological activity can be systemically expressed by SVRDF scales, which may be an useful structural expression methodology for the study of peptides QSAR.
为建立一种可应用于多肽定量构效关系（QSAR）研究的新氨基酸结构描述符，通过对150个氨基酸径向分布函数指数矩阵进行主成分分析，得到了一个新的描述符SVRDF。然后将该标度应用于由偏最小二乘回归构建的三组肽QSAR中。对于48种苦味二肽，得到的模型相关系数（R2(cum)）和交叉验证相关系数（Q2(cum)）分别为0.766和0.724；对于21种催产素类似物，分别为0.941和0.811；对于20种凝血酶原抑制剂，分别为0.996和0.919。令人满意的结果表明，SVRDF标度可以系统地表达与生物活性相关的信息，这可能是一种用于肽QSAR研究的有用结构表达方法。
