支持向量机在PPⅡ二级结构预测中的应用

PPⅡ二级结构是一种稀有的蛋白质结构类型。目前使用机器学习方法预测此二级结构的工作还比较少见。引入一种新的方法———支持向量机 (SVM)来预测PPII二级结构 ,并与神经网络方法进行了比较 ,结果表明 ,SVM方法在预测PPII结构上表现良好 ,预测精度达到 76 .5 2 %。     

自动编码器方法的蛋白质二级结构预测

蛋白质二级结构预测是进行蛋白质三级结构研究的重要基础,氨基酸的编码方式对二级结构预测有一定的影响。本文应用了一种新的组合编码方式,即将基团编码与位置特异性打分矩阵(PSSM)进行组合的编码方式。本文中提出的基团编码是针对氨基酸的一种新的编码方式,基团编码是根据氨基酸内部组成来进行编码的,由42位属性组成。本文选取位置特异性打分矩阵(PSSM)中的Blosum62进化矩阵和新的基团编码进行组合,形成新的编码方式。然后对CB513和25pdb两组数据分别进行实验。本文中将采用贝叶斯分类器与自动编码器两种方法来对这种新的编码方式进行实验,然后比较这两种方法得到的两组数据的结果。可以很明显的发现采用自动编码器的实验结果要比使用贝叶斯分类器的结果要高出1.65%。在本文的实验中,可以提取特征的自动编码器的预测准确率更好。     

基于支持向量机和贝叶斯方法的蛋白质四级结构分类研究

用支持向量机和贝叶斯两种方法对蛋白质四级结构进行分类研究。结果表明,基于支持向量机的分类结果最好,其l0CV检验的总分类精度、正样本正确预测率、Matthes相关系数和假阳性率分别为74．2％、84．6％、0．474、38．9％;基于贝叶斯的分类结果没有支持向量机的分类结果好,但其l0CV检验的假阳性率最低(15．9％)．这些结果说明同源寡聚蛋白质一级序列包含四级结构信息,同时特征向量的确表示了埋藏在缔合亚基作用部位接触表面的基本信息。     

基于序列特征的人类PolⅡ启动子理论预测

基于已知的人类PolII启动子序列数据,综合选取启动子序列内容和序列信号特征,构建启动子的支持向量机分类器．分别以启动子序列的6-mer频数作为离散源参数构建序列内容特征。同时选取24个位点的3-mer频数作为序列信号特征构建PWM,将所得到的两类参数输入支持向量机对人类启动子进行预测．用10折叠交叉检验和独立数据集来衡量算法的预测能力,相关系数指标达到95％以上,结果显示结合了支持向量机的离散增量算法能够有效的提高预测成功率,是进行真核生物启动子预测的一种很有效的方法．     

基于支持向量机的~(31)P磁共振波谱肝细胞癌诊断

支持向量机是在统计学习理论基础上发展起来的一种新的机器学习方法,在模式识别领域有着广泛的应用。利用基于支持向量机模型的31P磁共振波谱数据对肝脏进行分类,区别肝细胞癌,肝硬化和正常的肝组织。通过对基于多项式核函数和径向基核函数的支持向量机分类器进行比较,并且得到三种肝脏分类的识别率。实验表明基于31P磁共振波谱数据的支持向量机分类模型能够对活体肝脏进行诊断性的预测。     

基于具有多茎环特征GA-SVM的人类pre-miRNA的预测研究

MicroRNA（miRNA）是一类长度约为21 nt的非编码RNA,在动植物中发挥着重要而广泛的转录后调控作用. 现有的计算预测方法通常不能很好地识别具有多分枝茎环二级结构的pre miRNA.为进一步提高对pre miRNA的预测精度,本文在以往研究的基础上,新引用了一类多茎环生物学特征,将遗传算法（GA）与支持向量机（SVM）结合以进行特征选择,同时优化SVM分类器模型参数（c,g）,并对数据集的不平衡性进行处理,构造出新的分类器.本文采用人类pre miRNA作为研究数据集,通过5折交叉验证,实验结果显示,新的分类器能够有效地提高预测精度.     

基于离散增量的支持向量机算法预测蛋白质中四类简单超二级结构

蛋白质超二级结构预测是三级结构预测的一个非常重要的中间步骤。本文从蛋白质的一级序列出发,对5793个蛋白质中的四类简单超二级结构进行预测,以位点氨基酸为参数,采用3种片段截取方式,分别用离散增量算法预测的结果不理想,将组合的离散增量值作为特征参数输入支持向量机,取得了较好的预测结果,5交叉检验的平均预测总精度达到83.0%,Matthew’s相关系数在0.71以上。     

基于遗传算法特征选择的HBV再激活分类预测模型

探讨原发性肝癌患者精确放疗后乙型肝炎病毒(hepatitis b virus,HBV)再激活的危险特征和分类预测模型。提出基于遗传算法的特征选择方法,从原发性肝癌数据的初始特征集中选择HBV再激活的最优特征子集。建立贝叶斯和支持向量机的HBV再激活分类预测模型,并预测最优特征子集和初始特征集的分类性能。实验结果表明,基于遗传算法的特征选择提高了HBV再激活分类性能,最优特征子集的分类性能明显优于初始特征子集的分类性能。影响HBV再激活的最优特征子集包括:HBV DNA水平,肿瘤分期TNM,Child-Pugh,外放边界和全肝最大剂量。贝叶斯的分类准确性最高可达82.89%,支持向量机的分类准确性最高可达83.34%。     

机器学习方法在基因功能注释中的应用

目前,基于计算机数学方法对基因的功能注释已成为热点及挑战,其中以机器学习方法应用最为广泛。生物信息学家不断提出有效、快速、准确的机器学习方法用于基因功能的注释,极大促进了生物医学的发展。本文就关于机器学习方法在基因功能注释的应用与进展作一综述。主要介绍几种常用的方法,包括支持向量机、k近邻算法、决策树、随机森林、神经网络、马尔科夫随机场、logistic回归、聚类算法和贝叶斯分类器,并对目前机器学习方法应用于基因功能注释时如何选择数据源、如何改进算法以及如何提高预测性能上进行讨论。     

基于支持向量机的蛋白质同源寡聚体分类研究

基于支持向量机和贝叶斯方法,从蛋白质一级序列出发对蛋白质同源二聚体、同源三聚体、同源四聚体、同源六聚体进行分类研究,结果表明：基于支持向量机, 采用“一对多”和“一对一”策略, 其分类总精度分别为77.36%和93.43%, 分别比基于贝叶斯协方差判别法的分类总精度50.64%提高26.72和42.79个百分点.从而说明支持向量机可用于蛋白质同源寡聚体分类,且是一种非常有效的方法.对于多类蛋白质同源寡聚体分类,基于相同的机器学习方法（如支持向量机）,采用“一对一”策略比“一对多”效果好.同时亦表明蛋白质同源寡聚体一级序列包含四级结构信息.     

Fast structure alignment for protein databank searching.

A fast method is described for searching and analyzing the protein structure databank. It uses secondary structure followed by residue matching to compare protein structures and is developed from a previous structural alignment method based on dynamic programming. Linear representations of secondary structures are derived and their features compared to identify equivalent elements in two proteins. The secondary structure alignment then constrains the residue alignment, which compares only residues within aligned secondary structures and with similar buried areas and torsional angles. The initial secondary structure alignment improves accuracy and provides a means of filtering out unrelated proteins before the slower residue alignment stage. It is possible to search or sort the protein structure databank very quickly using just secondary structure comparisons. A search through 720 structures with a probe protein of 10 secondary structures required 1.7 CPU hours on a Sun 4/280. Alternatively, combined secondary structure and residue alignments, with a cutoff on the secondary structure score to remove pairs of unrelated proteins from further analysis, took 10.1 CPU hours. The method was applied in searches on different classes of proteins and to cluster a subset of the databank into structurally related groups. Relationships were consistent with known families of protein structure.     

Bayesian segmentation of protein secondary structure.

We present a novel method for predicting the secondary structure of a protein from its amino acid sequence. Most existing methods predict each position in turn based on a local window of residues, sliding this window along the length of the sequence. In contrast, we develop a probabilistic model of protein sequence/structure relationships in terms of structural segments, and formulate secondary structure prediction as a general Bayesian inference problem. A distinctive feature of our approach is the ability to develop explicit probabilistic models for alpha-helices, beta-strands, and other classes of secondary structure, incorporating experimentally and empirically observed aspects of protein structure such as helical capping signals, side chain correlations, and segment length distributions. Our model is Markovian in the segments, permitting efficient exact calculation of the posterior probability distribution over all possible segmentations of the sequence using dynamic programming. The optimal segmentation is computed and compared to a predictor based on marginal posterior modes, and the latter is shown to provide significant improvement in predictive accuracy. The marginalization procedure provides exact secondary structure probabilities at each sequence position, which are shown to be reliable estimates of prediction uncertainty. We apply this model to a database of 452 nonhomologous structures, achieving accuracies as high as the best currently available methods. We conclude by discussing an extension of this framework to model nonlocal interactions in protein structures, providing a possible direction for future improvements in secondary structure prediction accuracy.     

Predicting sumoylation site by feature selection method

The small ubiquitin-like modifier (SUMO) proteins are a kind of proteins that can be attached to a series of proteins. The sumoylation of protein is an important posttranslational modification. Thus, the prediction of the sumoylation site of a given protein is significant. Here we employed a combined method to perform this task. We predicted the sumoylation site of a protein by a two-staged procedure. At the first stage, whether a protein would be sumoylated was predicted; whereas at the second stage, the sumoylation sites of the protein were predicted if it was determined to be modified by SUMO at the first stage. At the first stage, we encoded a protein with protein families (PFAM) and trained the predictor with nearest network algorithm (NNA); at the second stage, we encoded nonapeptides (peptides that contain nine residues) of the protein containing the lysine residues, with Amino Acid Index, and trained the predictor with NNA. The predictor was tested by the k-fold cross-validation method. The highest accuracy of the second-staged predictor was 99.55% when 12 features were incorporated in the predictor. The corresponding Matthews Correlation Coefficient was 0.7952. These results indicate that the method is a promising tool to predict the sumoylation site of a protein. At last, the features used in the predictor are discussed. The software is available at request.     

Extended secondary structures in proteins

Supersecondary structures of proteins have been systematically searched and classified, but not enough attention has been devoted to such large edifices beyond the basic identification of secondary structures. The objective of the present study is to show that the association of secondary structures that share some of their backbone residues is a commonplace in globular proteins, and that such deeper fusion of secondary structures, namely extended secondary structures (ESSs), helps stabilize the original secondary structures and the resulting tertiary structures. For statistical purposes, a set of 163 proteins from the protein databank was randomly selected and a few specific cases are structurally analyzed and characterized in more detail. The results point that about 30% of the residues from each protein, on average, participate in ESS. Alternatively, for the specific cases considered, our results were based on the secondary structures produced after extensive Molecular Dynamics simulation of a protein–aqueous solvent system. Based on the very small width of the time distribution of the root mean squared deviations, between the ESS taken along the simulation and the ESS from the mean structure of the protein, for each ESS, we conclude that the ESSs significantly increase the conformational stability by forming very stable aggregates. The ubiquity and specificity of the ESS suggest that the role they play in the structure of proteins, including the domains formation, deserves to be thoroughly investigated.     

Protein packing: dependence on protein size, secondary structure and amino acid composition

We have used the occluded surface algorithm to estimate the packing of both buried and exposed amino acid residues in protein structures. This method works equally well for buried residues and solvent-exposed residues in contrast to the commonly used Voronoi method that works directly only on buried residues. The atomic packing of individual globular proteins may vary significantly from the average packing of a large data set of globular proteins. Here, we demonstrate that these variations in protein packing are due to a complex combination of protein size, secondary structure composition and amino acid composition. Differences in protein packing are conserved in protein families of similar structure despite significant sequence differences. This conclusion indicates that quality assessments of packing in protein structures should include a consideration of various parameters including the packing of known homologous proteins. Also, modeling of protein structures based on homologous templates should take into account the packing of the template protein structure.     

Secondary structure prediction and folding of globular protein: refolding of ferredoxin

The physicochemical mechanism of protein folding has been elucidated by the island model, describing a growth type of folding. The folding pathway is closely related with nucleation on the polypeptide chain and thus the formation of small local structures or secondary structures at the earliest stage of folding is essential to all following steps. The island model is applicable to any protein, but a high precision of secondary structure prediction is indispensable to folding simulation. The secondary structures formed at the earliest stage of folding are supposed to be of standard form, but they are usually deformed during the folding process, especially at the last stage, although the degree of deformation is different for each protein. Ferredoxin is an example of a protein having this property. According to X-ray investigation (1FDX), ferredoxin is not supposed to have secondary structures. However, if we assumed that in ferredoxin all the residues are in a coil state, we could not attain the correct structure similar to the native one. Further, we found that some parts of the chain are not flexible, suggesting the presence of secondary structures, in agreement with the recent PDB data (1DUR). Assuming standard secondary structures (-helices and -strands) at the nonflexible parts at the early stage of folding, and deforming these at the final stage, a structure similar to the native one was obtained. Another peculiarity of ferredoxin is the absence of disulfide bonds, in spite of its having eight cysteines. The reason cysteines do not form disulfide bonds became clear by applying the lampshade criterion, but more importantly, the two groups of cysteines are ready to make iron complexes, respectively, at a rather later stage of folding. The reason for poor prediction accuracy of secondary structure with conventional methods is discussed.     

Statistical model of amino acid code of protein secondary structure

In the previous paper (Shestopalov, 2003) we presented the amino acid code of protein secondary structure as a partial solution of the fundamental problem of the protein three-dimensional structure calculation from the amino acid sequence. Here a statistical model of the code is described. The model is based on the structural data from 2258 protein chains (417,112 amino acid residues used). 60 and 61% of the secondary structure, calculated using the model, coincide, respectively, with the observed secondary structure in the training subset and test subset (104 protein chains and 21,166 residues used). This is equal to the threshold value for all the secondary structure calculations, based on the models, where, similarly as here, only the nearest and middle-range interactions are considered. Therefore the constructed model can be applied for the protein structure prediction from the amino acid sequence, especially when additional information is used along with expert analysis, as in the most successful prediction methods. The model can be used for analysis of the secondary structure changes during protein folding by comparison of the calculated and observed secondary structures. The information about the conformationally invariant segments can serve for the simulation of the supersecondary structure formation. One can try to obtain and examine the protein subset, in which the calculated and observed secondary structures are very similar.     

ChSeq: A database of chameleon sequences

Chameleon sequences (ChSeqs) refer to sequence strings of identical amino acids that can adopt different conformations in protein structures. Researchers have detected and studied ChSeqs to understand the interplay between local and global interactions in protein structure formation. The different secondary structures adopted by one ChSeq challenge sequence‐based secondary structure predictors. With increasing numbers of available Protein Data Bank structures, we here identify a large set of ChSeqs ranging from 6 to 10 residues in length. The homologous ChSeqs discovered highlight the structural plasticity involved in biological function. When compared with previous studies, the set of unrelated ChSeqs found represents an about 20‐fold increase in the number of detected sequences, as well as an increase in the longest ChSeq length from 8 to 10 residues. We applied secondary structure predictors on our ChSeqs and found that methods based on a sequence profile outperformed methods based on a single sequence. For the unrelated ChSeqs, the evolutionary information provided by the sequence profile typically allows successful prediction of the prevailing secondary structure adopted in each protein family. Our dataset will facilitate future studies of ChSeqs, as well as interpretations of the interplay between local and nonlocal interactions. A user‐friendly web interface for this ChSeq database is available at prodata.swmed.edu/chseq .     

RBRDetector: Improved prediction of binding residues on RNA‐binding protein structures using complementary feature‐ and template‐based strategies

Computational prediction of RNA‐binding residues is helpful in uncovering the mechanisms underlying protein‐RNA interactions. Traditional algorithms individually applied feature‐ or template‐based prediction strategy to recognize these crucial residues, which could restrict their predictive power. To improve RNA‐binding residue prediction, herein we propose the first integrative algorithm termed RBRDetector (RNA‐Binding Residue Detector) by combining these two strategies. We developed a feature‐based approach that is an ensemble learning predictor comprising multiple structure‐based classifiers, in which well‐defined evolutionary and structural features in conjunction with sequential or structural microenvironment were used as the inputs of support vector machines. Meanwhile, we constructed a template‐based predictor to recognize the putative RNA‐binding regions by structurally aligning the query protein to the RNA‐binding proteins with known structures. The final RBRDetector algorithm is an ingenious fusion of our feature‐ and template‐based approaches based on a piecewise function. By validating our predictors with diverse types of structural data, including bound and unbound structures, native and simulated structures, and protein structures binding to different RNA functional groups, we consistently demonstrated that RBRDetector not only had clear advantages over its component methods, but also significantly outperformed the current state‐of‐the‐art algorithms. Nevertheless, the major limitation of our algorithm is that it performed relatively well on DNA‐binding proteins and thus incorrectly predicted the DNA‐binding regions as RNA‐binding interfaces. Finally, we implemented the RBRDetector algorithm as a user‐friendly web server, which is freely accessible at http://ibi.hzau.edu.cn/rbrdetector . Proteins 2014; 82:2455–2471. © 2014 Wiley Periodicals, Inc.     

Active site prediction for comparative model structures with thematics

THEMATICS (Theoretical Microscopic Titration Curves) is a simple, reliable computational predictor of the active sites of enzymes from structure. Our method, based on well-established Finite Difference Poisson-Boltzmann techniques, identifies the ionisable residues with anomalous predicted titration behavior. A cluster of two or more such perturbed residues is a very reliable predictor of the active site. The protein does not have to bear any resemblance in sequence or structure to any previously characterized protein, but the method does require the three-dimensional structure. We now present evidence that THEMATICS can also locate the active site in structures built by comparative modeling from similar structures. Results are given for a total of 21 sets of proteins, including 21 templates and 83 comparative model structures. Detailed results are presented for three sets of orthologous proteins (Triosephosphate isomerase, 6-Hydroxymethyl-7,8-dihydropterin pyrophosphokinase, and Aspartate aminotransferase) and for one set of human homologues of Aldose reductase with different functions. THEMATICS correctly locates the active site in the model structures. This suggests that the method can be applicable to a much larger set of proteins for which an experimentally determined structure is unavailable. With a few exceptions, the predicted active sites in the comparative model structures are similar to that of the corresponding template structure.