摘要
提出了用于表征分子内部化学微环境的结构描述子:原子电性作用矢量(AEIV),并将其应用于20个天然氨基酸103个13C原子核磁共振化学位移CS建模:δ=-190.514+7.352×ν1+63.998×ν2+49.252×ν3+39.678×ν4,取得优良效果.模型值、留一法(LOO)和留分法(LMO)交互校验的复相关系数分别为RMM=0.9660,RLOO=0.9577和RLMO=0.9577.
A novel method based on atomic electronegativity interaction vector (AEIV) was developed to describe chemical environment and atomic state in amino acids, and to predict {}+{13}C chemical shifts. One hundred and three {}+{13}C chemical shifts for twenty natural amino acids were calculated by the method. The precision of calculation is cross-validated using the chemical shifts calculated by established methods, including molecular modeling, the leave-one-out (LOO) method and the leave-molecule-out (LMO) method. The correlation coefficients (R) obtained are R-{MM}=0.966, R-{LOO}=0.958 and R-{LOO}={0.950,} respectively. The model developed was also applied to predict fifteen {}+{13}C NMR chemical shifts for two newly discovered natural amino acids.
出处
《波谱学杂志》
CAS
CSCD
北大核心
2005年第2期163-172,共10页
Chinese Journal of Magnetic Resonance
基金
霍英东基金(199876)
国家春晖计划教育部启动基金(199914/38)
重庆大学创新基金(200356)资助项目