摘要
以20种天然氨基酸的721个3D结构信息描述子经主成分分析,得出一种新的氨基酸描述子——SVTD(Scores Vector of Three Dimension Descriptors),将其应用于血管紧张素转化酶抑制剂与促凝血酶原酶抑制剂的结构表征,并应用偏最小二乘建模方法建立定量构效关系模型,建模复相关系数R2cum、交互检验复相关系数Q2cum分别为0.880、0.778和0.998、0.804。结果表明:SVTD描述子能够系统地表征与生物活性相关的结构信息,该描述子有望成为多肽定量构效关系研究中一种有效的结构表达方法。
In this work, a new descriptor, SVTD, was derived from principal components analysis of a matrix of 721 structural variables including 3D information for 20 coded amino acids. SVTD scales were then employed to construct QSAR models of angiotensin-converting enzyme inhibitors and thromboplastin inhibitors based on PLS regression. The correlation coefficients （R^2cum）, cross-validation correlation coefficients （Q^2cum） were 0. 880 and 0. 778 for 58 angiotensin-converting enzyme inhibitors; 0. 998 and 0. 804 for 20 thromboplastin inhibitors, respectively. Satisfactory results showed that information related to biological activity could be systemically expressed by SVTD scales,which may be an useful structural expression methodology for study on peptides QSARs.
出处
《分析科学学报》
CAS
CSCD
2008年第5期522-526,共5页
Journal of Analytical Science
基金
陕西省教育厅科研计划[08JK225]
陕西科技大学博士科研启动基金[BJ07-03
BJ07-04]
关键词
氨基酸
肽
定量构效关系
SVTD描述子
Amino acids
Peptides
Quantitative structure-activity relationship
SVTD descriptor
