基于岭回归和SVM的高维特征选择与肽QSAR建模

岭回归估计权重绝对值在一定程度上体现了对应特征作用大小, 据此发展了基于岭回归(RR)和支持向量机(SVM)的高维特征选择算法. 对苦味二肽(BTT)和细胞毒性T淋巴细胞(CTL)表位9 肽两个肽体系, 以氨基酸的531 个物理化学性质参数直接表征肽结构, 各获得1062、4779 个初始特征; 对训练集, 初始特征以岭回归排序后序贯引入, 当SVM留一法交叉测试(LOOCV)的均方误差(MSE)显著上扬时终止, 最后以多轮末尾淘汰进一步精筛, 分别获得7、18个物理化学意义明确的保留特征. 基于保留特征与支持向量回归(SVR), 对训练集建立定量构效关系(QSAR)模型, 预测独立测试集, 其拟合精度、留一法交叉测试精度、独立预测精度均优于现有文献报道结果. 新方法运行速度快, 选取的特征物理化学意义明确, 解释性强, 在肽、蛋白质定量构效关系建模等高维数据回归预测领域有较广泛应用前景.     

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。     

定量结构-活性/性质相关性中变量的选择——正交变换法与最优子集回归法的比较

定量结构－活性／性质相关性（QSAR／QSPR）研究的基本依据是化合物的性质与结构具有相关性,所以只要有方法描述化合物的结构（得到X）就可与化合物的性质（作为Y）建立起数学模型,并由引模型预测未知化合物。由化合物的结构可衍生（即描述）出诸多变量,从统计学出发,希望用尽可能少的变量来表征尽可能多的信息（如多元回归分析）。过多的变量不仅计算量大,从而可以导致所得的数学模型不稳定,使预测结果较差^[1],而且不同变量的组合所得结果可能差别很大,由此需要对变量进行压缩和选择。虽然变量的选择是一个非常费时和复杂的工作,但变量选择的好坏对数学模型的稳定性及准确性有致关重要的影响,从某种角度上讲,它能决定一项QSAR／QSPR研究的成败。最简单的选择变量的方法是穷举组合法,但此方法的计算量非常大,特别是当变量数较大时,该方法是实际上是不可行的,尽管用于变量选择的方法已有报道,但问题尚有待进一步研究。本文侧重比较了正交变换法与变量最优子集回归法,得到了很有启示性的结果。     

Local lazy regression: making use of the neighborhood to improve QSAR predictions

Traditional quantitative structure-activity relationship (QSAR) models aim to capture global structure-activity trends present in a data set. In many situations, there may be groups of molecules which exhibit a specific set of features which relate to their activity or inactivity. Such a group of features can be said to represent a local structure-activity relationship. Traditional QSAR models may not recognize such local relationships. In this work, we investigate the use of local lazy regression (LLR), which obtains a prediction for a query molecule using its local neighborhood, rather than considering the whole data set. This modeling approach is especially useful for very large data sets because no a priori model need be built. We applied the technique to three biological data sets. In the first case, the root-mean-square error (RMSE) for an external prediction set was 0.94 log units versus 0.92 log units for the global model. However, LLR was able to characterize a specific group of anomalous molecules with much better accuracy (0.64 log units versus 0.70 log units for the global model). For the second data set, the LLR technique resulted in a decrease in RMSE from 0.36 log units to 0.31 log units for the external prediction set. In the third case, we obtained an RMSE of 2.01 log units versus 2.16 log units for the global model. In all cases, LLR led to a few observations being poorly predicted compared to the global model. We present an analysis of why this was observed and possible improvements to the local regression approach.     

Quantitative structure-activity relationship studies of boron-containing dipeptide proteasome inhibitors using calculated mathematical descriptors

Topological indices (TIs) and atom pairs (APs) were used to develop quantitative structure-activity relationship (QSAR) models of a set of 58 dipeptide boronic acids which are potent inhibitors of proteasome and have found applications in the treatment of various types of cancers. Of the three linear regression methods used for QSAR development, viz., principal components regression (PCR), partial least square (PLS), and ridge regression (RR), the last method gave the most satisfactory models whereas the remaining two methods yielded poor models. RR results obtained in this paper using TIs and APs are comparable to the CoMFA and CoMSIA results reported in the literature with the same set of compounds.     

Determining the validity of a QSAR model--a classification approach

The determination of the validity of a QSAR model when applied to new compounds is an important concern in the field of QSAR and QSPR modeling. Various scoring techniques can be applied to specific types of models. We present a technique with which we can state whether a new compound will be well predicted by a previously built QSAR model. In this study we focus on linear regression models only, though the technique is general and could also be applied to other types of quantitative models. Our technique is based on a classification method that divides regression residuals from a previously generated model into a good class and bad class and then builds a classifier based on this division. The trained classifier is then used to determine the class of the residual for a new compound. We investigated the performance of a variety of classifiers, both linear and nonlinear. The technique was tested on two data sets from the literature and a hand built data set. The data sets selected covered both physical and biological properties and also presented the methodology with quantitative regression models of varying quality. The results indicate that this technique can determine whether a new compound will be well or poorly predicted with weighted success rates ranging from 73% to 94% for the best classifier.     

On further application of r as a metric for validation of QSAR models

Validation is a crucial aspect for quantitative structure–activity relationship (QSAR) model development. External validation is considered, in general, as the most conclusive proof of predictive capacity of a QSAR model. In the absence of truly external data set, external validation is usually performed on test set compounds, which are members of the original data set but not used in model development exercise. In the case of small data sets, QSAR researchers experience problem in model development due to the fact that the developed models may be less reliable on account of the small number of training set compounds and such models may also show poor external predictability because the models may not have captured all necessary features required for the particular structure–activity relationships. The present paper attempts to show that ‘true r_(LOO)’ statistic calculated based on the model derived from the undivided data set with application of variable selection strategy at each cycle of leave‐one‐out (LOO) validation may reflect external validation characteristics of the developed model thus obviating the requirement of splitting of the data set into training and test sets. This approach may be helpful in the case of small data sets as it uses all available data for model development and validation thus making the resulting model more reliable. Copyright © 2009 John Wiley & Sons, Ltd.