蛋白质二级结构预测概率图模型的改进

蛋白质二级结构与蛋白质三级结构及蛋白质功能密切相关,是生物信息学研究的热点,其中概率图模型隐马尔可夫算法(HMM)是该领域研究的重要工具。但是在实际应用中,存在着HMM训练下溢、不同训练集的效果差异较大及参数优化困难等问题。对预测蛋白质二级结构时HMM遇到的训练下溢问题提出了改进方案;首次提出8-状态HMM来预测蛋白质二级结构,并且将参数B改进成为包含状态转移信息的三维参数;为了改进最优HMM模型的确定方法,用每个样本分别对初始HMM模型进行训练,得到一系列新的模型,然后对这些新模型的参数求均值,将求得的均值作为最优模型的参数。这些改进方法提高了HMM预测蛋白质二级结构的准确率,为HMM的进一步优化打下良好的基础。     

THE蛋白的结构分析与功能预测

调用motif数据库、profile数据库和interproscan数据库,对THE蛋白进行了序列同源性分析和功能位点分析.结果表明,THE蛋白质是一种核蛋白,理论等电点为6.36,分子量为44 859 Dalton.应用多种相关软件对THE蛋白的二级结构和特殊结构进行了初步预测,结果显示:THE蛋白中存在α螺旋、β-折叠片和无规卷曲结构,有两个可形成跨膜结构的片段,不存在卷曲螺旋,无信号肽,也无线粒体定位信号.对THE蛋白进行序列同源性、结构域及功能位点预测,结果显示:THE蛋白与来自大鼠睾丸的Tes13-S、Fos13-L和几种假设蛋白有较高的相似性;THE蛋白存在次黄嘌呤核苷酸脱氢酶、嘌呤核苷酸还原酶结构域及PKC、酰胺化、豆蔻酸连接等功能位点,无糖基化位点.THE蛋白的结构分析与功能预测为该基因的功能研究提供了重要的依据.     

Hierarchically Clustered HMM for Protein Sequence Motif Extraction with Variable Length

Protein sequence motifs extraction is an important field of bioinformatics since its relevance to the structural analysis. Two major problems are related to this field：（1） searching the motifs within the same protein family; and（2） assuming a window size for the motifs search. This work proposes the Hierarchically Clustered Hidden Markov Model（HC-HMM） approach, which represents the behavior and structure of proteins in terms of a Hidden Markov Model chain and hierarchically clusters each chain by minimizing distance between two given chains＇ structure and behavior. It is well known that HMM can be utilized for clustering, however, methods for clustering on Hidden Markov Models themselves are rarely studied. In this paper, we developed a hierarchical clustering based algorithm for HMMs to discover protein sequence motifs that transcend family boundaries with no assumption on the length of the motif. This paper carefully examines the effectiveness of this approach for motif extraction on 2593 proteins that share no more than 25% sequence identity. Many interesting motifs are generated.Three example motifs generated by the HC-HMM approach are analyzed and visualized with their tertiary structure.We believe the proposed method provides a unique protein sequence motif extraction strategy. The related data mining fields using Hidden Markova Model may also benefit from this clustering on HMM themselves approach.     

应用型大学内部教学质量保障之组织体系的探索与实践

对比分析了国外与国内应用型大学内部教学质量保障之组织体系结构的差异，指出了我国大多数学校的教学质量保障之组织体系结构中的弱点，提出了应用型大学教学执行体系与教学质量评价体系既合作又分工的、“纵向分权”的组织结构，介绍了东北大学东软信息学院依据ABET国际标准和我国国家标准建立的东软教学质量保障体系，重点介绍了其组织结构，描述了各部门的职责和部门间的重要关系．还介绍了东软信息学院按照上述质量体系，全面开展的一系列内部教学质量评估与改善工作，重点介绍了课程质量评估．     

HMM in predicting protein secondary structure

We introduced a new method—duration Hidden Markov Model (dHMM) to predicate the secondary structure of Protein. In our study, we divide the basic second structure of protein into three parts: H (α-Helix), E (β-sheet) and O (others, include coil and turn). HMM is a kind of probabilistic model which more thinking of the interaction between adjacent amino acids (these interaction were represented by transmit probability), and we use genetic algorithm to determine the model parameters. After improving on the model and fixed on the parameters of the model, we write a program HMMPS. Our example shows that HMM is a nice method for protein secondary structure prediction. Foundation item: Supported by the National Natural Science Foundation of China (30170214) Biography: Huang Jing (1977-), female, Master candidate, research direction: bioinformatics.     

人SEPT9蛋白同源建模和活性位点分析

利用生物信息学在线软件预测了人SETP 9蛋白质的二级结构和模体信息,同时对其三级结构进行同源建模和模建结果质量评价,其次预测了该蛋白质的活性位点信息,旨在从蛋白质序列特征和分子结构水平理解其在人类生理病理过程中的作用.结果表明,模建的SEPT 9蛋白结构品质较高,具有7段α-螺旋和2组β-折叠结构,是一个典型的α/β类蛋白,表面呈弱正电势分布;人SEPT 9蛋白具有8个不同模体,可能参与不同生化反应或执行不同的功能.搜寻获得了人SEPT 9蛋白配基结合位点有10个,其中位点1可能是该蛋白的活性位点.这些研究结果对理解人SEPT 9蛋白功能以及配基结合位点定位非常重要,也为针对SEPT 9蛋白的分子对接和药物从头设计提供了理论基础.     

QoS保证的一种时间改进的制造服务流程优化方法

制造服务流程是一种基于业务流程的制造服务链,它有顺序、选择、循环、并行等4种基本结构,而循环能转化为顺序结构,因而选择结构和并行结构才是真正的分支结构。分支结构的各分支往往会有服务能力差异,这会导致:选择分支会因为概率分配不当将延误时间,而并行分支则会因此出现等待情况,这样,势必会影响制造服务流程整体的执行效率。为此,提出了QoS保证的一种时间改进的制造服务流程优化方法。构建了制造服务流程基本结构的属性计算方法,在分析了几种分支结构的时间与其他属性因子之间的影响关系后,基于QoS约束,构建了分支结构时间优化的分层分块线性规划模型,并设计了分层分块的线性优化算法。经实验,优化后的业务流程执行时间提高了5.4%,表明所建模型及其优化算法是有效且合理的,对云制造的应用具有积极意义。     

Assessment of protein models with three-dimensional profiles.

As methods for determining protein three-dimensional (3D) structure develop, a continuing problem is how to verify that the final protein model is correct. The revision of several protein models to correct errors has prompted the development of new criteria for judging the validity of X-ray and NMR structures, as well as the formation of energetic and empirical methods to evaluate the correctness of protein models. The challenge is to distinguish between a mistraced or wrongly folded model, and one that is basically correct, but not adequately refined. We show that an effective test of the accuracy of a 3D protein model is a comparison of the model to its own amino-acid sequence, using a 3D profile, computed from the atomic coordinates of the structure 3D profiles of correct protein structures match their own sequences with high scores. In contrast, 3D profiles for protein models known to be wrong score poorly. An incorrectly modelled segment in an otherwise correct structure can be identified by examining the profile score in a moving-window scan. The accuracy of a protein model can be assessed by its 3D profile, regardless of whether the model has been derived by X-ray, NMR or computational procedures.     

蛋白质结构样板库的构建及其总体特征

按照分辨率较高、同一性较低的标准 ,从 PDB库中选取高质量的蛋白质空间结构数据 ,构建了精度高于 0 .2 5nm、序列同一性低于 2 5%的样板数据库 SL CTBASE,其中有效残基数近 1 9万 .建库过程中 ,对 PDB原始数据文件进行了详细检验 ,发现其中包含有错误及可疑数据 .本文还统计了 SLCTBASE中主链键长、CA-CA距离、二面角的分布情况     

蛋白质拓扑结构预测的进一步讨论

利用信息论方法，找到了与蛋白质拓扑结构相关性较好的一些二级结构参数，确定了预测蛋白质拓扑结构的最佳参数、对α类，β类，α／β类蛋白制定了简洁的预测规则；对结果进行了讨论。     

基于语义提升HMM的语义标注

语义标注所用标签数目众多,训练数据更为稀疏,用HMM作语义标注面临参数估计不准的难题。不同于传统的解决数据稀疏方法,以《同义词词林》的层次式结构为依据,提出了利用语义层次的提升来改善HMM(hidden Markov model)中参数的估计质量;在算法实现中,采用选择受限策略来解决因语义提升而引起的模型辨别力下降问题。测试表明,在训练数据相对稀疏的情况下,适度调整模型的语义层次可大幅提高语义标注的精度,该方法表现出较好的可塑性。     

基于约束Laplacian分值的半监督特征选择算法

针对Laplacian分值法进行特征选择时过分依赖样本局部结构信息的不足，提出一种改进的基于约束Laplacian分值的半监督特征选择算法。该算法利用样本之间的cannot-link成对约束关系作为全局结构信息，在进行特征选择时，不仅能尽量保持局部结构信息，而且还尽量保持了全局的cannot-link约束关系。基于Yale和PIE(Fave pose,Illamination,Expression dadbase)人脸数据库的实验表明，该算法性能显著优于Laplacian分值法，与Fisher分值法和最新的约束分值法相当，且在稳定性方面优于后者。     

基于隐马尔可夫模型的DNA序列识别

利用隐马尔可夫模型训练中不同结构的DNA序列的L值分布范围不同的特点,对传统多类投票模型进行改进,提出一种优于传统算法的快速训练算法,该算法只需训练出一类隐马尔可夫模型参数.对DNA内含子和外显子序列进行识别,平均识别率达到了90.8%.与支持向量机相比,隐马尔可夫模型在解决多分类问题方面具有优势,不但计算时间少,而且识别率高.     

基于分层时序模型的步态识别算法

以隐马尔可夫模型和动态纹理模型为代表的动态贝叶斯网是描述步态序列的重要方法,但都存在一些不足之处.提出了一种新的动态贝叶斯网——分层时序模型,该方法采用分段线性逼近非线性和用各段的动态纹理模型作为隐状态,将隐马尔可夫模型和动态纹理模型做了结合,充分发挥了其优势.该方法在CMU Mobo步态数据库和CASIA步态数据库B上做了评估,结果充分显示了分层时序模型的高性能.     

同源建模方法预测蛋白质突变结构的适用性分析

通过从Protein Data Bank(PDB)结构数据库中提取单氨基酸突变的晶体结构,构建了一组无冗余的测试数据集,对目前应用最广泛的两款同源建模预测软件(SWISS-MODEL和MODELLER)进行了测试分析,发现它们对蛋白质的整体结构预测效果良好,均方根偏差小于0.5埃(RMSD0.5),但在突变导致结构显著变化(RMSD1.5)的情况下却均不能得到准确结果.分类统计显示,发生在蛋白质结构内部和极性氨基酸之间的突变结构变化小,两款软件预测效果较好(RMSD1.0).突变导致结构显著变化的可能性不高(5%),但它对蛋白质功能的影响不可忽视,因此应用同源建模方法对于蛋白质突变的模拟并不完全适用,还需要开发新方法来提高准确性.     

一种蛋白质结构同源建模的DNA算法

将一种新的智能计算方法——DNA计算引入到蛋白质结构预测中,试图建立当蛋白质结构与母板结构相似度比较低的情况下的蛋白质结构比较模型的DNA计算方法。将氨基酸序列中的一个残基的可能构型映射为一段DNA序列,将一个蛋白质结构预测问题转化成一个边赋权的图的最大权团问题,结合最大权团问题的DNA计算模型,建立蛋白质预测问题的DNA算法,并通过仿真实验说明了算法的有效性。     

一种基于关联图的蛋白质结构预测改进算法

采用一种改进的COMAR （Contact Map Reconstruction）算法求解基于关联图的蛋白质结构预测问题.根据蛋白质关联图和先验知识，并以半随机的方式生成距离信息，根据距离信息得到蛋白质的坐标，并通过坐标修正和摄动，使得重构结构的关联图与给定的关联图相一致.结果表明，阈值较大的关联图所重建的结构较好，与原COMAR算法相比，在相同的迭代次数下，改进的COMAR算法的精度较高.     

Protein dispensability and rate of evolution.

If protein evolution is due in large part to slightly deleterious amino acid substitutions, then the rate of evolution should be greater in proteins that contribute less to individual fitness. The rationale for this prediction is that relatively dispensable proteins should be subject to weaker purifying selection, and should therefore accumulate mildly deleterious substitutions more rapidly. Although this argument was presented over twenty years ago, and is fundamental to many applications of evolutionary theory, the prediction has proved difficult to confirm. In fact, a recent study showed that essential mouse genes do not evolve more slowly than non-essential ones. Thus, although a variety of factors influencing the rate of protein evolution have been supported by extensive sequence analysis, the relationship between protein dispensability and evolutionary rate has remained unconfirmed. Here we use the results from a highly parallel growth assay of single gene deletions in yeast to assess protein dispensability, which we relate to evolutionary rate estimates that are based on comparisons of sequences drawn from twenty-one fully annotated genomes. Our analysis reveals a highly significant relationship between protein dispensability and evolutionary rate, and explains why this relationship is not detectable by categorical comparison of essential versus non-essential proteins. The relationship is highly conserved, so that protein dispensability in yeast is also predictive of evolutionary rate in a nematode worm.     

Complex Sequence Quality Evaluation Model for Concealed Water Supply Pipeline of Prefabricated Construction

The accurate assessment of the quality can materially affect the safety and life of the prefabricated construction. In this paper, we studied the overall module of concealed water supply pipeline, and a new complex sequence quality evaluation model was established. By analyzing the small evaluation index elements of the model, the weight of the larger evaluation index could be deduced. The total quality evaluation score of the overall module could be obtained by calculating the weight value of the evaluation index, and then the quality standard of the overall module could be obtained.     

基于模体中心结构的蛋白质复合物识别算法

基于蛋白质相互作用的网络有明显模块化特征, 其对预测蛋白质功能、解释特定的生物进程具有重要作用, 网络模体是复杂网络演化的 重要拓扑结构, 其代表了复杂系统中的重要功能单元, 具有进化保守性的特性, 提出一种新的基于网络模体为核心节点组的蛋白质复合物识别算法. 该算法根据蛋白质相互作用网络的拓扑特性, 将模体作为蛋白质复合物的中心结构体, 并基于中心结构体进行二层节点扩充, 能准确有效地识别蛋白质复合物. 并且将复合物二维网络进行三维转化, 从而更直观清晰地展示复合物的结构体特征. 实验结果验证了该算法的有效性及可行性.