基于完备集预测芳香类化合物~(13)C核磁共振波谱
|
摘要
在建立分子定量结构-性能关系(QSPR)模型过程中,需要挑选分子结构参数子集.但目前还没有统一的参数挑选方法,参数集所用的分子参数可多可少,带有主观性.参数完备集具备哲学之美,不多一个元素,也不少一个元素.本文基于完备集建立61种芳香类化合物695个~(13)C核磁共振(NMR)化学位移(δC)QSPR模型.完备集分子参数基于PBE1PBE/6-311G(2d,2p)量子化学方法计算得到.采用Duplex算法对数据集进行划分,并用支持向量机(SVM)结合粒子群优化算法(PSO)建立~(13)C NMR化学位移的QSPR模型.所建的2个SVM模型对整个数据集预测的均方根误差(rms)均为2.4ppm,小于广义回归神经网络(GRNN)模型预测结果;此结果与文献报道值相比也是精确的.结果表明,应用完备集建立~(13)C NMR化学位移SVM预测模型是成功的;且为QSPR建模提供了新的参数集挑选方法.
Choosing the best set of descriptors is a key step for developing quantitative structure-property relationship(QSPR)models.However,nowadays there is no general method to reveal descriptor's importance.Thus choosing the best set of descriptors is subjective and different descriptor set may be obtained for the same case.A complete set of descriptors means that the set is perfectness since there is no any element redundant or need to be added.Here we report the application of complete sets of descriptors calculated with PBE1 PBE/6-311 G(2 d,2 p)approach to develop QSPR models for 695 ~(13)C NMR chemical shifts(δCparameters)of carbon atoms in 61 aromatic compounds.Duplex algorithm is used to split the data sets.Two QSPR models forδCparameters are developed with support vector machine(SVM)algorithm,by applying the particle swarm optimization(PSO)technique to optimize SVM parameters.Both of the two SVM models have root mean square(rms)errors of 2.4 ppm for the total set of 695δCparameters,which are less than the errors from two general regression neural network(GRNN)models.Compared with previous QSPR models for ~(13)C NMR chemical shifts,the prediction results are accurate,which suggest that applying complete sets of descriptors for SVM models is successful.This study provides a new selection method of descriptor set in QSPR modeling.
Choosing the best set of descriptors is a key step for developing quantitative structure-property relationship(QSPR)models.However,nowadays there is no general method to reveal descriptor's importance.Thus choosing the best set of descriptors is subjective and different descriptor set may be obtained for the same case.A complete set of descriptors means that the set is perfectness since there is no any element redundant or need to be added.Here we report the application of complete sets of descriptors calculated with PBE1 PBE/6-311 G(2 d,2 p)approach to develop QSPR models for 695 ~(13)C NMR chemical shifts(δCparameters)of carbon atoms in 61 aromatic compounds.Duplex algorithm is used to split the data sets.Two QSPR models forδCparameters are developed with support vector machine(SVM)algorithm,by applying the particle swarm optimization(PSO)technique to optimize SVM parameters.Both of the two SVM models have root mean square(rms)errors of 2.4 ppm for the total set of 695δCparameters,which are less than the errors from two general regression neural network(GRNN)models.Compared with previous QSPR models for ~(13)C NMR chemical shifts,the prediction results are accurate,which suggest that applying complete sets of descriptors for SVM models is successful.This study provides a new selection method of descriptor set in QSPR modeling.
引文
[1]Verma R P,Hansch C.Use of 13 C NMR Chemical Shift as QSAR/QSPR Descriptor[J].Chem.Rev.,2011,111(4):2865-2899.
[2]VilkováM,Mal’uckáLU,Imrich J.Prediction by 13CNMR of Regioselectivity in 1,3-dipolar Cycloadditions of Acridin-9-yldipolarophiles[J].Magn.Reson.Chem.,2016,54(1):8-16.
[3]March R E,Burns D C,Ellis D A.Empirically Predicted 13 C NMR Chemical Shifts for 8-hydroxyflavone Starting from 7,8,4’-trihydroxyflavone and from 7,8-dihydroxyflavone[J].Magn.Reson.Chem.,2008,46(7):680-682.
[4]Xin D,Sader C A,Chaudhary O,et al.Development of a 13 C NMR Chemical Shift Prediction Procedure U-sing B3LYP/cc-pVDZ and Empirically Derived Systematic Error Correction Terms:A Computational Small Molecule Structure Elucidation Method[J].J.Org.Chem.,2017,82(10):5135-5145.
[5]Iron M A.Evaluation of the Factors Impacting the Accuracy of 13 C NMR Chemical Shift Predictions using Density Functional Theory-The Advantage of LongRange Corrected Functionals[J].J.Chem.Theory Comput.,2017,13(11):5798-5819.
[6]Baldridge K K,Siegel J S.Quantum Chemical Prediction of the 13 C NMR Shifts in Alkyl and Chlorocorannulenes:Correction of Chlorine Effects[J].Theor.Chem.Acc.,2008,120(1-3):95-106.
[7]Shaghaghi H,Fathi F,Ebrahimi H P,et al.Quantitative Prediction of 13 C NMR Chemical Shifts in Solvent Using PCM-ONIOM Method and Optimally Selected Wave Function[J].Concept.Magn.Reson.A,2013,42A(1):1-13.
[8]Tulyabaev A R,Kiryanov II,Samigullin I S,et al.Are There Reliable DFT Approaches for 13 C NMRChemical shift Predictions of Fullerene C60Derivatives[J].Int.J.Quantum Chem.,2017,117(1):7-14.
[9]Tong J,Liu S,Zhou P,et al.Quantitative Structure Spectroscopy Relationships of Carbon-13 Nuclear Magnetic Resonance Chemical Shifts of Steroids[J].J.Mol.Graph.Model.,2007,26(1):86-92.
[10]Blinov K A,Smurnyy Y D,Churanova T S,et al.Development of a Fast and Accurate Method of 13 C NMRChemical Shift Prediction[J].Chemomet.Intellig.Lab.Sys.,2009,97(1):91-97.
[11]Zhou L P,Sun L L,Yu Y,et al.Prediction of Carbon-13 NMR Chemical Shift of Alkanes with Rooted Path Vector[J].J.Mol.Graph.Model.,2006,25(3):333-339.
[12]Liu X,Ren Y,Zhou P,et al.Prediction of Protein 13CαNMR Chemical Shifts Using a Combination Scheme of Statistical Modeling and Quantum-mechanical Analysis[J].J.Mol.Struct.,2011,995(1-3):163-172.
[13]Kiryanov II,Mukminov F H,Tulyabaev A R,et al.Prediction of 13 C NMR Chemical Shifts by Artificial Neural Network.I.Partial Charge Model as Atomic Descriptor[J].Chemomet.Intellig.Lab.Sys.,2016,158:62-68.
[14]龚运淮,丁立生.天然产物核磁共振碳谱分析[M].昆明:云南科技出版社,2005.
[15]Yu X,Wang Y,Yang H,et al.Prediction of the Binding Affinity of Aptamers Against the Influenza Virus[J].SAR QSAR Environ.Res.,2019,30(1):51-62.
[2]VilkováM,Mal’uckáLU,Imrich J.Prediction by 13CNMR of Regioselectivity in 1,3-dipolar Cycloadditions of Acridin-9-yldipolarophiles[J].Magn.Reson.Chem.,2016,54(1):8-16.
[3]March R E,Burns D C,Ellis D A.Empirically Predicted 13 C NMR Chemical Shifts for 8-hydroxyflavone Starting from 7,8,4’-trihydroxyflavone and from 7,8-dihydroxyflavone[J].Magn.Reson.Chem.,2008,46(7):680-682.
[4]Xin D,Sader C A,Chaudhary O,et al.Development of a 13 C NMR Chemical Shift Prediction Procedure U-sing B3LYP/cc-pVDZ and Empirically Derived Systematic Error Correction Terms:A Computational Small Molecule Structure Elucidation Method[J].J.Org.Chem.,2017,82(10):5135-5145.
[5]Iron M A.Evaluation of the Factors Impacting the Accuracy of 13 C NMR Chemical Shift Predictions using Density Functional Theory-The Advantage of LongRange Corrected Functionals[J].J.Chem.Theory Comput.,2017,13(11):5798-5819.
[6]Baldridge K K,Siegel J S.Quantum Chemical Prediction of the 13 C NMR Shifts in Alkyl and Chlorocorannulenes:Correction of Chlorine Effects[J].Theor.Chem.Acc.,2008,120(1-3):95-106.
[7]Shaghaghi H,Fathi F,Ebrahimi H P,et al.Quantitative Prediction of 13 C NMR Chemical Shifts in Solvent Using PCM-ONIOM Method and Optimally Selected Wave Function[J].Concept.Magn.Reson.A,2013,42A(1):1-13.
[8]Tulyabaev A R,Kiryanov II,Samigullin I S,et al.Are There Reliable DFT Approaches for 13 C NMRChemical shift Predictions of Fullerene C60Derivatives[J].Int.J.Quantum Chem.,2017,117(1):7-14.
[9]Tong J,Liu S,Zhou P,et al.Quantitative Structure Spectroscopy Relationships of Carbon-13 Nuclear Magnetic Resonance Chemical Shifts of Steroids[J].J.Mol.Graph.Model.,2007,26(1):86-92.
[10]Blinov K A,Smurnyy Y D,Churanova T S,et al.Development of a Fast and Accurate Method of 13 C NMRChemical Shift Prediction[J].Chemomet.Intellig.Lab.Sys.,2009,97(1):91-97.
[11]Zhou L P,Sun L L,Yu Y,et al.Prediction of Carbon-13 NMR Chemical Shift of Alkanes with Rooted Path Vector[J].J.Mol.Graph.Model.,2006,25(3):333-339.
[12]Liu X,Ren Y,Zhou P,et al.Prediction of Protein 13CαNMR Chemical Shifts Using a Combination Scheme of Statistical Modeling and Quantum-mechanical Analysis[J].J.Mol.Struct.,2011,995(1-3):163-172.
[13]Kiryanov II,Mukminov F H,Tulyabaev A R,et al.Prediction of 13 C NMR Chemical Shifts by Artificial Neural Network.I.Partial Charge Model as Atomic Descriptor[J].Chemomet.Intellig.Lab.Sys.,2016,158:62-68.
[14]龚运淮,丁立生.天然产物核磁共振碳谱分析[M].昆明:云南科技出版社,2005.
[15]Yu X,Wang Y,Yang H,et al.Prediction of the Binding Affinity of Aptamers Against the Influenza Virus[J].SAR QSAR Environ.Res.,2019,30(1):51-62.