AutoMotif Server for prediction of phosphorylation sites in proteins using support vector machine: 2007 update

We present here the recent update of AutoMotif Server (AMS 2.0) that predicts post-translational modification sites in protein sequences. The support vector machine (SVM) algorithm was trained on data gathered in 2007 from various sets of proteins containing experimentally verified chemical modifications of proteins. Short sequence segments around a modification site were dissected from a parent protein, and represented in the training set as binary or profile vectors. The updated efficiency of the SVM classification for each type of modification and the predictive power of both representations were estimated using leave-one-out tests for model of general phosphorylation and for modifications catalyzed by several specific protein kinases. The accuracy of the method was improved in comparison to the previous version of the service (Plewczynski et al., “AutoMotif server: prediction of single residue post-translational modifications in proteins”, Bioinformatics 21: 2525–7, 2005). The precision of the updated version reached over 90% for selected types of phosphorylation and was optimized in trade of lower recall value of the classification model. The AutoMotif Server version 2007 is freely available at . Additionally, the reference dataset for optimization of prediction of phosphorylation sites, collected from the UniProtKB was also provided and can be accessed at .     

A shaking bioreactor equipped with twin ceramic membranes for acetic acid production using Acetobacter pasteurianus

A shaking bioreactor system with twin internal ceramic membranes was developed for effective perfusion culture and applied to the continuous production of acetic acid using Acetobacter pasteurianus. The system makes it possible to carry out the back-washing of the membrane without stopping the continuous operation because one membrane can be washed by medium feed flow while another membrane provides filtration of the broth by the simple switching of the medium and the broth flow direction. The medium flow through the membrane could successfully wash the surface of the membrane thereby effectively maintaining the filtration ability. By using the system, continuous operation of more than 800 h was achieved and the maximum acetic acid productivity reached 13.4 g l^–1 h^–1 using air enriched with 40% O₂.     

Support vector machines for prediction of protein subcellular location by incorporating quasi-sequence-order effect.

Support Vector Machine (SVM), which is one class of learning machines, was applied to predict the subcellular location of proteins by incorporating the quasi-sequence-order effect (Chou [2000] Biochem. Biophys. Res. Commun. 278:477-483). In this study, the proteins are classified into the following 12 groups: (1) chloroplast, (2) cytoplasm, (3) cytoskeleton, (4) endoplasmic reticulum, (5) extracellular, (6) Golgi apparatus, (7) lysosome, (8) mitochondria, (9) nucleus, (10) peroxisome, (11) plasma membrane, and (12) vacuole, which account for most organelles and subcellular compartments in an animal or plant cell. Examinations for self-consistency and jackknife testing of the SVMs method were conducted for three sets consisting of 1,911, 2,044, and 2,191 proteins. The correct rates for self-consistency and the jackknife test values achieved with these protein sets were 94 and 83% for 1,911 proteins, 92 and 78% for 2,044 proteins, and 89 and 75% for 2,191 proteins, respectively. Furthermore, tests for correct prediction rates were undertaken with three independent testing datasets containing 2,148 proteins, 2,417 proteins, and 2,494 proteins producing values of 84, 77, and 74%, respectively.     

基于优化核参数支持向量机的意识任务分类

根据支持向量机的基本原理,给出一种推广误差上界估计判据,并利用该判据进行最优核参数的自动选取。对三种不同意识任务的脑电信号进行多变量自回归模型参数估计,作为意识任务的特征向量,利用支持向量机进行训练和分类测试。分类结果表明,优化核参数的支持向量机分类器取得了最佳的分类效果,分类正确率明显高于径向基函数神经网络。     

Continuous measurements of electrical conductivity in growing media using a modified suction probe: Initial calibration and potential usefulness

A electrical conductivity (EC) cell introduced into a porous ceramic cup was developed to continuously sample the solution and measure EC from different growing media. Application of pressure head creates a continuous flow of solution from the growing media tested, into the ceramic cup, and through the EC cell. Continuous recording of the EC was achieved by connecting the EC meter to a data logger. Using two different pressure heads (–5 and –15 kPa) allowed us to observe differences in the EC of the solution extracted that resulted from the different moisture retention of each growing media. After a maximum period of 24 h extracting the solution from different growing media, EC values obtained with the probe were compared with those obtained using paste extracts, saturated with either deionized water or nutrient solution. Results show EC values obtained using the probe with a –15 kPa pressure head were closer to values of EC measured in saturated extracts made with nutrient solution. Using a –5 kPa pressure head, EC values with probe were lower than those obtained by extracts done with nutrient solution, but higher than EC values from saturated extracts made with deionized water. Simultaneous measurements of matric tension and EC show the effect of pressure heads applied in the probe on the water content of growing media. This technique is not destructive (the sampling of growing media is not necessary) and it is possible to obtain EC measurements of solutions continuously. This method of measuring water and salt content of the root environment has potential applications in the greenhouse production of pot plants. Measurements obtained with this method may lead to new information on nutrient uptake by plants and the development of new strategies of managing fertility and irrigation of horticultural crops.     

基于氨基酸约化和统计特征的蛋白质亚细胞定位预测

蛋白质亚细胞定位预测对蛋白质的功能、相互作用及调控机制的研究具有重要意义。本文基于物化性质和结构性质对氨基酸的约化,描述序列局部和全局信息的"组成"、"转换"和"分布"特征,并利用氨基酸亲疏水性的数值统计特征,提出了一种新的蛋白质特征表示方法(NSBH)。分别使用三种分类器KNN、SVM及BP神经网络进行蛋白质亚细胞定位预测,比较了几种方法和特征融合方法的预测结果,显示融合特征表示及结合SVM分类器时能够达到更好的预测准确率。同时,还详细讨论了不同参数对实验结果的影响,具体的实验及比较结果显示了该方法的有效性。     

用ID-SVM预测蛋白质的ATP结合位点

从蛋白质序列出发,对经Dr．G．P．S．Raghava整理和使用过的168条非冗余的ATP与蛋白质结合氨基酸序列进行分段,对ATP与蛋白质结合位点进行了统计分析。在此基础上,利用20种氨基酸的亲疏水性将20种氨基酸约化为6类。以氨基酸组分和6类亲疏水紧邻为参数,用多样性增量（ID）方法将氨基酸组分和6类亲疏水紧邻降维并将降维后的特征参数输入支持向量机中运算,本文运算结果显示用氨基酸组分ID值和6类亲疏水紧邻ID值共同作为特征参数结果最优,在七交叉检验下的预测总精度达到了99．67％,相关系数达到0．9934,好于前人的预测结果。     

基于近邻成分分析算法的原发性肝癌精确放疗后HBV再激活分类预测

原发性肝癌(PLC)患者在精确放疗后乙型肝炎病毒(HBV)再激活是一种常见并发症,及时的预测防护能降低发病率、死亡率。研究表明:多余的特征变量会影响HBV再激活的预测精度。通过提出基于近邻成分分析(NCA)的特征选择方法找出HBV再激活的危险因素及特征组合。之后分别建立经Bayes优化前后的支持向量机模型(SVM)对这些关键特征子集及初始特征集进行分类预测。实验结果表:明HBV DNA水平、KPS评分、分割方式、外放边界、V25、肿瘤分期TNM、ChildPugh等都是影响HBV再激活的危险因素。其中经NCA特征选择之后发现的V25是在乙型肝炎病毒再激活研究中首次提出的危险因素。10折交叉验证下特征组合HBV DNA水平、外放边界、V25的预测精度高达86.11%。支持向量机分类器可以很好的应用于乙型肝炎病毒再激活的研究,特征选择后的关键特征组合具有更优越的分类性能。     

鸟类形态特征的同质性比较

鸟类作为脊椎动物中的主要类群之一,其内部演化关系一直是动物学研究的重点。形态特征及分子特征在推断鸟类内部类群间的系统发育关系时被广泛应用,但长久以来受制于同质性的影响,不同形态特征难以得到一致的系统发育关系。随着基因组测序技术及分析方法的快速发展,基于大量DNA序列数据所建的物种树成为领域内普遍认可的鸟类系统发育关系。在此基础之上,重新挖掘形态特征的研究价值成为可能。本研究基于Prum发表于2015年的鸟类科级水平系统发育树,比较了不同形态特征同质性水平的差异,讨论了形态特征对于提高分子特征系统发育信号及分子树支持率的作用。研究发现,不同种类形态特征的同质性水平存在显著的差异,以一致性指数为标准,杂项特征(包括神经、肌腱及肠道的特征)的得分显著高于骨学特征和肌学特征(P < 0.01),颅骨特征的得分显著高于非颅骨、躯干及腿部特征的得分(P < 0.05);加入形态特征能够显著提高分子特征的系统发育信号(P < 0.05),也提高了分子树的支持率。综上所述,不同的形态特征在同质性水平的得分上具有显著差异,在分子特征上加入形态特征既能够提高分子特征的系统发育信号,也能够提高分子树的支持率。     

一种基于支持向量机的自适应肿瘤分类检测算法

根据肿瘤分类检测模型的特点,提出了一种新的算法,该算法结合使用了基因选择和数据抽取的有效方法,并在此基础上使用支持向量机对基因表达数据进行分类或者检测。其中乳腺癌的分类交叉验证结果由88．46％提高到100．0％,急性白血病的也由71．05％提高至100．0％。实验结果说明了这一方法的有效性,为在大量的基因表达数据中提高检测癌症的准确性提出了一种比较通用的方法。