一种新的氨基酸描述符及其在肽QSAR中的应用

从20种天然氨基酸的41个randic molecular profiles非零描述符、44个eigenvalue based indices非零描述符和47个walk and path counts非零描述符分别进行主成分分析,得出一种新的氨基酸描述符-SVREW。将其应用于血管紧张素转化酶(ACE)抑制二肽和ACE抑制三肽、苦味二肽和苦味四肽、后叶催产素类似物、HLA-A*0201限制性CTL表位肽的结构表征,应用多元线性回归(MLR)建立定量构效关系模型,同时采用内部与外部双重验证的方法验证模型的稳定性。所建ACE抑制二肽、ACE抑制三肽、苦味二肽、苦味四肽、后叶催产素类似物、HLA-A*0201限制性CTL表位肽的模型复相关系数(R2cum)分别为0.994,0.797,0.948,0.878,0.686,0.720;留一法交互校验复相关系数(R2cv)分别为0.955,0.859,0.879,0.958,0.796,0.843;外部样本校验相关系数(Q2ext)分别为0.990,0.954,0.890,0.950,0.748,0.773。经研究表明SVREW描述符用于肽分子结构表征所建模型的稳定性与预测能力均较好,有望成为多肽定量构效关系研究中一种有效的结构表征方法,可对新药物的发现和研究提供指导。     

神经网络在ACE抑制肽QSAR中的应用研

通过对天然氨基酸的457种物化性质参数进行主成分分析后得到SVHEHS描述符,用该描述符分别对血管紧张素转化酶(ACE)抑制二肽、三肽、四肽进行表征,并建立了肽结构与活性的神经网络模型。ACE抑制二肽神经网络模型的相关系数、交叉验证相关系数、均方根误差和外部验证相关系数分别为0.946、0.951、0.249、0.852,三肽模型分别为0.973、0.945、0.135、0.813,四肽模型分别为0.915、0.879、0.250、0.814。由此表明SVHEHS描述符结合神经网络对ACE抑制肽的建模效果及模型预测能力均较理想,在此基础上进一步通过平均影响值(Mean impact value,MIV)法确定了显著影响各类肽活性的结构因素,从而为新的强活性ACE抑制肽的分子设计提供了理论基础。     

基于氨基酸描述符SVHEHS的ACE抑制肽QSAR研究

收集20种天然氨基酸的457种理化性质,按照疏水、电性特征、氢键贡献和立体特征分类后,对它们分别进行主成分分析(Principal component analysis,PCA),得到一个新的氨基酸残基结构描述符SVHEHS.用该描述符分别对血管紧张素转化酶(AngiotensinⅠconverting enzyme,ACE)抑制二肽、三肽、四肽进行序列表征,并用来与生物活性建立偏最小二乘(Partial least square regression,PLS)模型.ACE抑制二肽、三肽、四肽模型的相关系数、交叉验证相关系数、 均方根误差、外部验证相关系数分别为0.607,0.507,0.587,0.783;0.852,0.813,0.232,0.839;1,1,0,0.935.由此说明,采用SVHEHS描述符建立的PLS模型拟合、预测能力均较好,可用于血管紧张素转化酶抑制肽的定量构效关系研究.     

基于新型氨基酸描述符SVICE对肽类药物进行QSAR研究

采用20种天然氨基酸的47个information indices描述符、33个connectivity indices描述符和44个eigenvalue-based indices描述符分别进行主成分分析,得出一种新的氨基酸描述符-SVICE.将其分别对三肽血管收缩素转化酶(ACE)、抗菌十八肽(AMP)、苦味活性二肽(BTT)序列表征后,建立结构与活性的SMR-MLR模型,并采用内外部双重验证的方法检验模型的稳定性.所建模型相关统计参量如下:复相关系数(Rcum2)、留一法(LOO)交互校验复相关系数(RCV2)和外部样本校验复相关系数(Qext2)分别为0.988,0.964,0.985;0.990,0.970和0.855;0.949,0.887,0.830.结果表明,运用SVICE描述符建立的MLR模型拟合、预测能力均较好,能较好解释肽类药物的活性与结构间的关系从而为新的强活性肽类药物的分子设计和改造提供了指导.     

ACE抑制剂的QSAR研究与分子设计

从20种天然氨基酸的41个randic molecular profiles、44个eigenvalue based indices和47个walk and path counts非零描述符分别进行主成分分析,得出一种新的氨基酸描述符——SVREW.将其应用于血管紧张素转化酶抑制三肽结构表征,应用多元线性回归(MLR)及偏最小二乘(PLS)建立定量构效关系模型,同时采用内部与外部双重验证的方法验证模型的稳定性.所建模型复相关系数(Rcum2)、留一法(LOO)交互校验相关系数(Rcv2)和外部样本校验相关系数(Qext2)分别为MLR(0.994,0.974,0.991),P LS(0.949,0.886,0.898).然后利用此多元线性回归方程设计出一系列血管紧张素转化酶抑制三肽化合物并预测了其活性,并且应用分子对接验证所设计药物的合理性.经研究表明SVREW描述符应用于ACE三肽结构表征所建模型的稳定性与预测能力均较好,有望成为多肽定量构效关系研究中一种有效的结构表征方法,并对新药物的发现和研究提供指导.     

基于不同建模方法的ACE抑制肽QSAR比较研究

以自组建的血管紧张素转化酶(Angiotensin I-converting enzyme)抑制肽库为研究对象,采用氨基酸描述符SVHEHS(Scores vector of hydrophobic,electronic,hydrogen bonds and steric properties)对各肽样本进行结构表征后,进行自交叉协方差(Auto cross covariances,ACC)处理,并分别利用多元线性回归(Multiple linear regression,MLR)、偏最小二乘(Partial least square regression,PLS)、人工神经网络(Artificial neural networks,ANN)3种建模方法进行ACE抑制肽QSAR建模。结果显示,所得MLR、PLS与ANN模型的相关系数(Correlation coefficient,R2)分别为0.744、0.862、0.958,留一交叉验证相关系数(Leave-one-out cross-validated correlation coefficient,Q2LOO)分别为0.532、0.829、0.948,外部验证复相关系数(External validated correlation coefficient,Q2ext)分别为0.567、0.632、0.634。因此,SVHEHS结合上述3种建模方法均适用于ACE抑制肽的QSAR研究,其中ANN的建模效果最优。     

基于逐步非线性回归的血管紧张素转化酶抑制肽QSAR建模

线性特征选择方法可提升定量构效关系(QSAR)模型的预测能力,但易忽略特征(理化属性)与分子活性间的非线性关系。本文提出基于支持向量回归(SVR)的逐步非线性回归(SSNR)特征选择算法并用于降血压药物血管紧张素转化酶(ACE)抑制肽的QSAR研究。首先以具有不同背景的5组分子描述符分别表征肽序列,以SSNR实施特征选择,再通过智能一致性模型(ICM)对各组描述符对应子模型的预测活性进行加权整合,获得最终活性预测值。在ACE抑制二肽与三肽两个数据上的应用结果表明,SSNR获得的特征子集结合ICM策略可有效提升模型预测能力(二肽的平均Q■为0.675±0.002,三肽为0.663±0.013),优于遗传算法-偏最小二乘(0.538±0.049、0.599±0.047)与逐步线性回归(0.583±0.041、0.675±0.010)。最后基于抑制活性已知肽序列预测所有活性未知肽的活性,分析了高活性肽及其氨基酸偏好性,为人工合成潜在高活性ACE抑制肽提供可能的序列组合。     

红葡萄酒香气成分色谱保留时间模拟

采用化合物非氢原子固有特征值和非氢原子之间的电性作用为结构描述符,对红葡萄酒香气成分中的65个化合物进行了结构表征。通过多元线性回归(MLR)和逐步回归(SMR)方法建立了该类化合物结构-色谱保留时间关系(QSRR)模型。模型的复相关系数(R)为0.907,标准偏差(SD)为4.507。用留一法(LOO)交互检验对模型进行了评价,得到的复相关系数(RCV)为0.849,标准偏差(SDCV)为5.656。结果表明,采用的分子结构描述符能够较好地表现该类化合物结构特征,所建模型具有较好的预测能力和稳定性。     

ACE抑制三肽与ACE相互作用的分子机制

4种食源性三肽IRP(Ile-Arg-Pro),IKP(Ile-Lys-Pro),GRP(Gly-Arg-Pro),IRA(Ile-ArgAla)的ACE抑制活性已得到实验证实,但其与ACE的相互作用模式与分子机制尚不清楚,本研究采用柔性分子对接方法解决这一问题。分子对接结果表明:4种三肽与ACE有相似的作用模式,氢键、亲水、疏水、静电等作用力共同对三肽与ACE的结合存在贡献,但以氢键作用为主;ACE分子中Lys511,His513,Tyr520,Tyr523等氨基酸残基为其与肽结合的重要结合位点;ACE抑制三肽中氮端氨基和碳端羧基对其抑制活性影响显著,其中氮端氨基的作用更为重要。通过以上分子机理研究可为开发强活性ACE抑制肽提供理论指导。     

基于QSAR模型的有机磷化合物毒性预测

随着有机磷化合物(OPs)的广泛应用,其在越来越多的环境介质中被检测出来。大多数OPs具有毒性,但人们缺乏快速且有效的预测手段来对毒性进行评估。本文将结合E-Dragon软件计算的分子描述符,采用不同的QSAR模型对36个OPs的毒性进行预测。文中采用后退法作为描述符筛选方法,以均方根误差(RMSE)作为评价标准,共找到14个对线性核函数支持向量机(SVM)模型贡献较大的描述符;在最终得到的SVM模型交叉验证结果中,计算值与实际值的相关系数为0. 913,均方根误差为0. 388;外部测试验证结果中,平均相对误差为9. 10%。此外,采用多元线性回归(MLR)、人工神经网络(ANN)以及偏最小二乘回归(PLS)模型对OPs的毒性进行预测,交叉验证结果显示,三个模型的计算值与实际值的相关系数分别为0. 878、0. 686与0. 620,没有SVM模型的预测能力好。因此采用线性核函数的SVM模型对OPs进行毒性预测是一个行之有效的方法。     

QSAR and Docking Studies of Thiazolidine-4-carboxylic Acid Derivatives as Neuraminidase Inhibitors

In order to understand the chemical-biological interactions governing their activities toward neuraminidase(NA), QSAR models of 28 thiazolidine-4-carboxylic acid derivatives with inhibitory influenza A virus were developed. Here a quantitative structure activity relationship(QSAR) model was built by three-dimensional holographic atomic vector field(3 D-HoVAIF) and multiple linear regression(MLR). The estimation stability and prediction ability of the model were strictly analyzed by both internal and external validations. The correlation coefficient(R2) of established MLR model was 0.984, and the cross-validated correlation coefficient(Q2) of MLR model was 0.947. Furthermore, the cross-validated correlation coefficient for the test set(Qext2) was 0.967. The binding mode pattern of the compounds to the binding site of integrase enzyme was confirmed by docking studies. The results of present study indicated that this model can aid in designing more potent neuraminidase inhibitors.     

三维全息原子场作用矢量用于吡咯类抗艾滋病药物QSAR研究

采用三维全息原子场作用矢量(3D-HoVAIF)对32个吡咯类抗艾滋病药物进行结构参数化表征,并与其活性建立定量构效关系。分别采用多元线性回归(MLR)和偏最小二乘(PLS)进行建模,建模的复相关系数(R2cum)、交互校验复相关系数(Q2cum)和模型的标准偏差(SD)分别为R2cum=0.914、Q2cum=0.812、SD=0.236(MLR);R2cum=0.836、Q2cum=0.719、SD=0.314(PLS),结果均优于文献值(R2cum=0.667,Q2cum=0.581,SD=0.420)。所建模型具有良好的稳定性和预测能力,表明3D-HoVAIF能够较好地表征该类分子的结构,值得进一步推广应用。     

Structural Characterization and Acute Toxicity Prediction of Substituted Aromatic Compounds by Using Molecular Vertexes Correlative Index

A molecular structural characterization (MSC) method called molecular vertexes correlative index (MVCI) was used to describe the structures of 30 substituted aromatic compounds. Through multiple linear regression (MLR) and stepwise multiple regression (SMR), a quantitative structure-toxicity relationship (QSTR) model with 4 variables was obtained. The correlation coefficient (R) of the model was 0.9467. Through partial least-squares regression (PLS), another QSTR model with 5 principal components was obtained. The correlation coefficient (R) of the model was 0.9518. Both models were evaluated by performing the cross-validation with the leave-one-out (LOO) procedure and the Cross-Validation (CV) correlation coefficients (RCV) were 0.9208 and 0.9214, respectively. The results suggested good stability and predictability of the models, and the molecular vertexes correlative index could successfully describe the structures of the substituted aromatic compounds.     

Novel Rearrangement of Small Peptides in Electrospray Ionization Tandem Mass Spectrometry

Electrospray ionization mass spectrometry (ESI-MS) is a powerful method for sequencing peptides. A novel fragmentation pattern with the loss of a neutral fragment of 45 Da was observed with the dipeptides, tripeptides,tetrapeptides and pentapeptides containing phenylalanine or histidine residues. A novel rearrangement reaction with the extrusion of a formamide piece was studied and the rearrangement mechanism was proposed and confirmed by deuterium labeling experiments with ESI-MS^n and high-resolution mass spectrometry. These findings are potentially helpful in identifying the specific sequence pattern in the peptide sequencing.     

2D-QSAR Using MLR and 3D-QSAR Using CoMSIA Studies on the Toxicity of Aromatic Hydrocarbons on Larval Sinonvaculina Constricta

Aromatic hydrocarbons,one of the persistent organic pollutants(POPs),has been usually found in mussels,accumulated for their hard mobility and activities in harbours and estuaries.In this study,based on the 96 hr-LC50 of 12 aromatic hydrocarbons with larval sinonvaculina constricta,three-dimensional quantitative structure-activity relationship(3D-QSAR) technique:comparative molecular similarity indices analysis(CoMSIA) and 2D-QSAR technique:multiple linear regression(MLR) were described to obtain more detailed insight into the structure-activity relationships between the molecular structure and bio-activity.The results show the MLR model based on density functional theory(DFT) calculation carried out at the B3LYP/6-311** level with Gaussian 03 program yielded a very good correlation with a coefficient squared R2 of 0.716 and a cross-validated Q2 of 0.874.The dipole moment and enthalpy,as the thermodynamic parameters,were two important factors influencing pLC50.Correspondingly,CoMSIA based on the partial least-squares(PLS) methodology with steric,electrostatic,hydrophobic,H-bond donor and acceptor fields contributing simultaneously were employed and the values of R2 and the cross validation with leave-One-Out(LOO) Q2LOO were 0.585 and 0.990,respectively,which reveals the structure features,such as the electronegative substituent(nitro-group),hydrophobic groups(the benzene ring) and H-bond(nitro-group),related to the toxicity.The results of 2D-QSAR employing MLR model and 3D-QSAR employing CoMSIA model provide the useful information for predicting the toxicity of other aromatic hydrocarbons by comparing the molecular structures of similar compounds.