基于支持向量机的药物诱导磷脂质病预测模型

本文应用一种组合遗传算法和共轭梯度法的支持向量机(GA-CG-SVM)方法建立了药物诱导磷脂质病分类预测模型.首先对描述符进行了优化,选出了19个描述符用于模型的构建,所建模型对训练集的预测准确率为81.6%,对测试集的预测精度为87.5%,说明所建SVM分类模型不仅能正确预测训练集药物诱导的磷脂质病,也对其他化合物具...     

Studies on monomer reactivity ratios. II. A charge-transfer model

A charge-transfer model is proposed for the treatment of monomer reactivity ratios in free-radical bulk polymerization. The procedure involves the assignment of three parameters to each monomer, which can be interpreted as being related to the energies of the highest occupied monomer orbital, the singly occupied radical orbital, and the lowest lying virtual orbital of the monomer. Parameters are found for 17 monomers and computed reactivity ratios for a large number of copolymer systems are tabulated and compared with experiment. Similarities of the present model and the electronegativity scheme are discussed.     

New simple procedures for the prediction of monomer reactivity ratios and transfer constants in radical polymerisation

The Q-e and Patterns of Reactivity schemes for the quantitative treatment of the reactions of polymer radicals both suffer from drawbacks which seriously limit their usefulness. Two revised versions of the Patterns of Reactivity scheme, in which all the necessary parameters are derived from experimental data, are summarised here. Both new schemes provide significantly more accurate assessments of monomer reactivity ratios and transfer constants than do their predecessors.     

Studies on monomer reactivity ratios. I. An electronegativity model

A model based on an electronegativity scheme is proposed for treatment of monomer reactivity ratios in free-radical bulk copolymerization. Values for each monomer are assigned to three parameters: a relative localization (or resonance stabilization) energy, a radical electronegativity, and a monomer electronegativity. Parameters for 17 monomers are given and calculated reactivity ratios are tabulated for a large number of copolymerizations. Agreement with experiment is usually obtained to within experimental error except for systems involving acrylonitrile. Computed parameters are rationalized on the basis of molecular orbital theory.     

Analysis of the linear methods for determining copolymerization reactivity ratios. II. A critical reexamination of cationic monomer reactivity ratios

A thorough examination of some cationic copolymerization systems by a new method has shown that many published r values have to be corrected significantly and that some are erroneous and meaningless, because for these systems the conventional copolymer compositions equation does not hold. Available information in regard of cationic copolymerizations has been treated in terms of three classes: (a) Systems in which the conventional copolymer composition equation adequately describes the copolymerization mechanism and previous authors justifiably used the two parameter model to calculate reactivity ratios. Our results show that the discrepancy between published r values and the more precise values obtained in this work is about ±23%. (b) Systems in which the approximations implicit in the conventional copolymer composition equation do not hold and the calculated r values are erroneous and misleading. Monomer pairs comprising monomers of significantly different reactivities belong to this class indicating that in copolymerizations in general and in cationic copolymerizations in particular a strong cast system exists, i.e., copolymerization can readily occur within the cast (between monomers of similar reactivities); however, only with difficulty if at all between casts (between monomers of differing reactivities). (c) Systems in which the use of the copolymer composition equation is completely unjustified, the calculated r values are meaningless and in some cases the existence of true copolymers is questioned.     

Reactivity in radical polymerisation: An improved method for the prediction of monomer reactivity ratios and transfer constants

Half a century after the Q-e scheme was promulgated, a substantially improved method for predicting monomer reactivity ratios and transfer constants has been developed from the Patterns of Reactivity scheme. It is based on the same premise as its predecessors, that reactivity is controlled partly by thermodynamic and partly by polar factors, but the new treatment yields values much closer, on average, to the reported experimental data.     

A real‐time model based on optimized least squares support vector machine for industrial polypropylene melt index prediction

The accurate and reliable real‐time prediction of melt index (MI) is indispensable in quality control of the industrial propylene polymerization (PP) processes. This paper presents a real‐time soft sensor based on optimized least squares support vector machine (LSSVM) for MI prediction. First, the hybrid continuous ant colony differential evolution algorithm (HACDE) is proposed to optimize the parameters of LSSVM. Then, considering the complexity and nondeterminacy of PP plant, an online correcting strategy (OCS) is adopted to update the modeling data and to revise the model's parameters adaptively. Thus, the real‐time prediction model, HACDE‐OCS‐LSSVM, is obtained. Based on the data from a real PP plant, the models of HACDE‐LSSVM, DE‐LSSVM and LSSVM are also developed for comparison. The research results show that the proposed real‐time model achieves a good performance in the practical industrial MI prediction process. Copyright © 2016 John Wiley & Sons, Ltd.     

The cross-species prediction of bacterial promoters using a support vector machine

Due to degeneracy of the observed binding sites, the in silico prediction of bacterial sigma(70)-like promoters remains a challenging problem. A large number of sigma(70)-like promoters has been biologically identified in only two species, Escherichia coli and Bacillus subtilis. In this paper we investigate the issues that arise when searching for promoters in other species using an ensemble of SVM classifiers trained on E. coli promoters. DNA sequences are represented using a tagged mismatch string kernel. The major benefit of our approach is that it does not require a prior definition of the typical -35 and -10 hexamers. This gives the SVM classifiers the freedom to discover other features relevant to the prediction of promoters. We use our approach to predict sigma(A) promoters in B. subtilis and sigma(66) promoters in Chlamydia trachomatis. We extended the analysis to identify specific regulatory features of gene sets in C. trachomatis having different expression profiles. We found a strong -35 hexamer and TGN/-10 associated with a set of early expressed genes. Our analysis highlights the advantage of using TSS-PREDICT as a starting point for predicting promoters in species where few are known.     

蛋白质组质谱分析中基于串并联支持向量机的肽段色谱保留时间预测方法

基于质谱的大规模蛋白质鉴定中,在线液相色谱分离发挥了重要作用。色谱保留时间(retention time,RT)是肽段鉴定和定量的重要信息。由于整个色谱分析运行时间中,流动相中的有机相采用了非线性浓度曲线以及样品中肽段之间的相互影响等因素,基于肽段序列的RT预测还存在精度不高、模型推广性能差等问题。本文提出了一种基于串并联支持向量机(serial and parallel support vector machine,SP-SVM)的RT预测方法,能够表征洗脱过程中有机相浓度的非线性变化和肽段之间的相互影响,显著提高了肽段保留时间预测的精度。利用复杂样本数据集验证结果表明,预测RT和实验RT之间的决定系数达到了0.95,超过95%的鉴定肽段的RT预测误差范围小于总运行时间的20%,超过70%的鉴定肽段的RT预测误差范围小于总运行时间的10%。本文提出的模型的性能达到了目前已知的最好水平。     

Comparative recalculation of the monomer reactivity ratios in copolymerization for styrene and methyl methacrylate

The monomer reactivity ratios (MMRs) in radical copolymerization for styrene and methyl methacrylate were recalculated by five different methods using literature copolymerization data. The use of approximate 95% confidence limits and their visual inspection helps to separate possibly biased copolymer composition data. The recalculated mean MRR values were r₁ (styrene) = 0.501 ± 0.031 and r₂ = 0.472 ± 0.031. The results of the linear least-squares calculation procedures seldom approach the quality of the nonlinear least-squares analysis according to the method of Tidwell and Mortimer.     

Application of the monomer reactivity ratios to the kinetic‐model discrimination and the solvent‐effect determination for the styrene/acrylonitrile monomer system

The impact of reactivity ratios determined with the Nelder and Mead simplex method on the kinetic‐model discrimination and the solvent‐effect determination for the styrene/acrylonitrile monomer system was investigated. For the monomer system, the penultimate unit effect was inversely proportional to the polarity of the solvent: acetonitrile < N,N‐dimethylformamide < methyl ethyl ketone < toluene. Quantitatively, the penultimate unit effect could be correlated with an absolute value of the difference between the standard deviation of the reactivity ratios determined for the terminal and penultimate models. By application of the F test, the penultimate model was justified for copolymerization in toluene. The conclusion was less certain for polymerization in methyl ethyl ketone. With a scanning procedure based on the simplex method, it was found that an equivalent representation of the copolymer‐composition data could be achieved with multiple sets of penultimate‐model reactivity ratios. However, the relationship between the triad‐sequence distribution and copolymer composition depended on the reactivity‐ratio set chosen for the microstructure determination. The microstructure calculated with the penultimate‐model reactivity ratios determined with the simplex method from the initial guess (r₁₁ = r₁, r₂₁ = 1/r₂, r₂₂ = r₂, r₁₂ = 1/r₁) did not obey the general “bootstrap effect” rule. This observation still requires some theoretical interpretation. © 2000 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 38: 846–854, 2000     

A new method of estimating monomer reactivity ratios in copolymerization by linear regression methods

Sequential gas-liquid chromotographic analysis of the reaction mixture throughout a copolymerization reaction in conjuction with the improved curve-fitting I (integrated form) method, which accounts for measurements errors in both variables, allows accurate estimation of the monomer reactivity ratios. In this article an alternative method is presented for estimating r values in copolymerization with linear regression only, which is especially suited to cases in which one or two of the r values is close to 1. In these cases the improved curve-fitting I method tends to converge slowly because of the numerical instability of the integrated copolymerization equation. The use of the new method is illustrated for the estimation of the r values for ethylene and vinyl acatate in benzene at 35 kg/cm² and 62°C. The linear regression method was also tried on other copolymerizations and the results are compared with those obtained from the improved curve-fitting I method. The limits of the applicability of the linear regression method were determined by simulated sequential sampling experiments. It appears that the new method is applicable when the product of the r values is between 0.001 and 2, provided both monomer conversions are large enough compared with the measurements error.     

A feature vector integration approach for a generalized support vector machine pairwise homology algorithm

Due to the exponential growth of sequenced genomes, the need to quickly provide accurate annotation for existing and new sequences is paramount to facilitate biological research. Current sequence comparison approaches fail to detect homologous relationships when sequence similarity is low. Support vector machine (SVM) algorithms approach this problem by transforming all proteins into a feature space of equal dimension based on protein properties, such as sequence similarity scores against a basis set of proteins or motifs. This multivariate representation of the protein space is then used to build a classifier specific to a pre-defined protein family. However, this approach is not well suited to large-scale annotation. We have developed a SVM approach that formulates remote homology as a single classifier that answers the pairwise comparison problem by integrating the two feature vectors for a pair of sequences into a single vector representation that can be used to build a classifier that separates sequence pairs into homologs and non-homologs. This pairwise SVM approach significantly improves the task of remote homology detection on the benchmark dataset, quantified as the area under the receiver operating characteristic curve; 0.97 versus 0.73 and 0.70 for PSI-BLAST and Basic Local Alignment Search Tool (BLAST), respectively.     

X-ray photoelectron spectroscopy: A new method for determination of copolymer composition,and monomer reactivity ratios and monomer sequence distribution therefrom

Copolymer compositions and reactivity ratios for the radical copolymerization of styrene with acrylonitrile have been determined by x-ray photoelectron spectroscopy (XPS). The results obtained by this technique were confirmed by elemental as well as ¹H-NMR (nuclear magnetic resonance) analysis. The monomer sequence distributions have also been calculated utilizing the monomer reactivity ratio values obtained by three different techniques viz., XPS, ¹H-NMR, and elemental analysis. The agreement between the monomer sequence distributions in the copolymer chain by these methods is very satisfactory. © 1997 John Wiley & Sons, Inc. J Polym Sci B: Polym Phys 35: 2049–2056, 1997     

Some observations on monomer reactivity ratios in aqueous and non-aqueous media

Reactivity ratios have been determined for free radical copolymerization of n.butyl acrylate (BA) with N-vinyl-2-pyrrolidone (VP) at 65°C in bulk and in the presence of water. For 0 to 30% by weight of water in the feed, the value of r_VP was sensibly constant (∼ 0.03 ± 0.03). The value of r_BA also underwent virtually no change (∼ 0.90 ± 0.05) at water contents from 0 to 22% by weight but increased to 3.7 at 25% water and to 6.3 at 30% water. These findings are not amenable to satisfactory explanation but one aspect of possible relevance is proposed tentatively.     

Development of a sugar-binding residue prediction system from protein sequences using support vector machine

Several methods have been proposed for protein–sugar binding site prediction using machine learning algorithms. However, they are not effective to learn various properties of binding site residues caused by various interactions between proteins and sugars. In this study, we classified sugars into acidic and nonacidic sugars and showed that their binding sites have different amino acid occurrence frequencies. By using this result, we developed sugar-binding residue predictors dedicated to the two classes of sugars: an acid sugar binding predictor and a nonacidic sugar binding predictor. We also developed a combination predictor which combines the results of the two predictors. We showed that when a sugar is known to be an acidic sugar, the acidic sugar binding predictor achieves the best performance, and showed that when a sugar is known to be a nonacidic sugar or is not known to be either of the two classes, the combination predictor achieves the best performance. Our method uses only amino acid sequences for prediction. Support vector machine was used as a machine learning algorithm and the position-specific scoring matrix created by the position-specific iterative basic local alignment search tool was used as the feature vector. We evaluated the performance of the predictors using five-fold cross-validation. We have launched our system, as an open source freeware tool on the GitHub repository (https://doi.org/10.5281/zenodo.61513).     

药物分子胎盘屏障渗透的支持向量回归模型

药物分子透过胎盘屏障进入胎儿体内可能对胎儿造成不良影响,在临床及药物研发中评估药物分子的胎盘屏障渗透能力,对于保护受孕期的妇女以及胎儿,都有着十分重要的作用.本文使用了组合的遗传算法-共轭梯度-支持向量回归(GA-CG-SvR)方法,建立了一个药物分子胎盘屏障渗透的预测模型.对于所建模型,使用5重交叉验证和独立测试集方...