蛋白质功能位点预测

在 IBM-PC 机上开发了蛋白质功能位点预测软件:PROSITE.根据 EMBL发布于激光光盘上的蛋白质功能位点氨基酸片段的保守模式数据库,对给定的蛋白质序列,可按19类443个氨基酸保守模式来探测蛋白质的所属家族,各种功能区的位置和活性部位等性质,通过52个序列的验证结果和 SWWISS 蛋白质数据库相一致.此外该软件还具有操作灵活,多种输入输出方式等特点。     

面向蛋白质功能位点识别的机器学习平台构建

有关蛋白质功能的研究是解析生命奥秘的基础,机器学习技术在该领域已有广泛应用。利用支持向量机(support vectormachine,SVM)方法,构建一个预测蛋白质功能位点的通用平台。该平台先提取非同源蛋白质序列,再对这些序列进行特征编码(包括序列的基本信息、物化特征、结构信息及序列保守性特征等),以编码好的样本作为训练数据,利用SVM进行训练,得到敏感性、特异性、Matthew相关系数、准确率及ROC曲线等评价指标,反复测试,得到评价指标最优的SVM模型后,便可以用来预测蛋白质序列上的功能位点。该平台除了应用在预测蛋白质功能位点之外,还可以应用于疾病相关单核苷酸多态性(SNP)预测分析、预测蛋白质结构域分析、生物分子间的相互作用等。     

杨树蛋白质磷酸化位点预测

以小黑杨磷酸化蛋白质组为研究对象,用人工神经网络表达丝氨酸、苏氨酸等残基位点的磷酸化与氨基酸序列的结构特征之间的非线性关系,建立了BP人工神经网络模型,并用磷酸化数据对所建模型进行训练和分析,得适宜的结构为21×16∶8∶4,拟合准确度为90%,Acc、Sn、Sp、MCC分别为78%、89%、67%、0.57,对比分析结果表明,所建模型具有较强的预测能力。     

利用序列保守模体和局部构象信息预测转录因子结合位点

转录因子结合位点的计算预测是研究基因转录调控的重要环节,但常用的位置特异得分矩阵方法预测特异性偏低.通过深入分析结合位点的生物特征,提出了一种综合利用序列保守模体和局部构象信息的结合位点预测方法,以极大相关得分矩阵作为保守模体的描述模型,并根据二苷参数模型计算位点序列的局部构象,将两类信息得分组合为多维特征向量,在二次判别分析的框架下进行训练和滑动预测.预测过程中还引入了位置信息量以优化似然得分和过滤备选结果.针对大肠杆菌CRP和Fis结合位点数据的留一法测试结果表明,描述模型的改进和多种信息的融合能有效地改善预测方法的性能,大幅度提高特异性.     

核酸-蛋白质结合能在剪切位点识别中的应用

把最大信息原理应用到核酸序列的保守位点分析中。利用最大信息原理,推导出了核酸和蛋白质特异性结合时的结合能表达式,并且估计了和蛋白质发生相互作用的核酸序列上的位点范围。为了检验此理论是否较为成功地反映了核酸和蛋白质结合时的实际情况,把它应用到基因内含子剪切位点的识别中,识别结果达到了较高的敏感性和特异性,这说明利用最大信息原理推导结合能表达式及估计核酸序列上参与反应的位点范围的理论是较为成功的。此研究结果一方面有助于核酸和蛋白质相互作用的理解,另一方面,也有助于和蛋白质发生相互作用的各种核酸序列的计算机识别研究。     

固有无序蛋白与蛋白质相互作用位点残基特征分析

本文对固有无序蛋白(IDPs)与其他蛋白质相互作用位点残基特征进行了研究.首先在数据库中选出满足条件的109条IDPs蛋白质链及与其他配体蛋白形成的299个IDPs-蛋白质复合物,然后提取复合物中作为相互作用位点的IDPs-蛋白质残基.这109条IDPs链中共含有50 031个氨基酸残基,其中处于作用位点的残基有4 822个.通过分析发现,20种氨基酸在形成IDPs-蛋白质相互作用位点残基时具有不同的倾向性,根据形成作用位点残基的倾向性,20种氨基酸可分成三大类:倾向型氨基酸(ILE、LEU、ARG、PHE、TYR、MET、TRP)、中间型氨基酸(GLN、GLU、THR、LYS、VAL、ASP、HIS)、非倾向型氨基酸(PRO、SER、GLY、ALA、ASN、CYS).研究结果还进一步表明,不同氨基酸在有序区域与无序区域形成IDPs-蛋白质作用位点残基的倾向性不同.其中,氨基酸TRP、LEU、ILE、CYS在有序和无序区域形成作用位点残基的差异性尤为明显,而氨基酸GLU、PHE、HIS、ALA则基本没有多大差别.对IDPs-蛋白质相互作用位点残基理化特征进行分析发现:疏水性强、侧链净电荷量较少、极性较小、溶剂可及性表面积较大、侧链体积较大、极化率较大的氨基酸比较倾向于形成作用位点残基.主成分分析结果显示,残基的极化率、侧链体积和溶剂可及表面积对作用位点残基影响最大.     

基于最优分割位点的蛋白质亚细胞位点预测方法

蛋白质的亚细胞位点信息有助于我们了解蛋白质的功能以及它们之间的相互作用,同时还可以为新药物的研发提供帮助。目前普遍采用的亚细胞位点预测方法主要是基于N端分选信号或氨基酸组分特征,但研究表明,单纯基于N端分选信号或氨基酸组分的方法都会丢失序列的序信息。为了克服此缺陷,本文提出了一种基于最优分割位点的蛋白质亚细胞位点预测方法。首先,把每条蛋白质序列分割为N端、中间和C端三部分,然后在每个子序列和整条序列中分别提取氨基酸组分、双肽组分和物理化学性质,最后我们把这些特征融合起来作为整条序列的特征。通过夹克刀检验,该方法在NNPSL数据集上得到的总体精度分别是87.8%和92.1%。     

基于氨基酸序列预测蛋白质功能性点突变位点

突变是研究蛋白质结构和功能的重要方法。点突变实验中,突变位点的选择随机性大,若能对突变后蛋白质功能是否发生变化做出预测,将大大减少实验的盲目性。为此,作者设计了一个基于信号处理的单点替换突变预测模型,对序列上每个位点所有可能的氨基酸替换的效果进行估计。使用蛋白质突变数据库(Protein Mutant Database,PMD)里的11个蛋白共2600多个点突变的数据集,对以上模型进行了验证。结果表明正确率高达81.2%,并且推荐出的替换选择位点仅占所有可能替换突变的3.1%。在体外定点突变实验中,使用本模型推荐的高可能性功能突变位点将有助于提高实验的成功率。该模型使用蛋白质的氨基酸序列信息,特别是对未知结构的蛋白质同样适用。然而,由于缺乏足够的突变实验数据,本模型的应用仍需进一步完善和验证。     

MiyabenolC和KobophenolA与雌激素受体的结合位点

iyabenolC (MiyC)和kobophenolA (KobA)是两种新型的植物雌激素。为了探讨MiyC和KobA与雌激素受体 (ER)的结合部位 ,运用计算机辅助分子模拟构建它们与ER结合的空间模型 ,找出结合位点 ,设计ER的两个突变体M1ER(ERM517AG52 1D)和M2ER(ERE353GR394 G) ;运用PCR技术将ER与MiyC或KobA的结合位点进行突变 ;运用报告基因检测实验 ,检测MiyC和KobA对突变的ER是否具有激活功能。结果显示MiyC激活M1ER使之促下游基因转录的作用下降 ,KobA对M1ER无激活作用 ;MiyC和KobA对M2ER无激活作用。以上结果显示MiyC和KobA与ER的结合位点可能为ER的Glu353 、Arg394 、Met517和Gly52 1。     

真核蛋白质的亚细胞位点预测研究进展

研究真核蛋白质的亚细胞位点是了解真核蛋白质功能,深入研究蛋白质相关信号通路内在机制的基础。同时,可以为了解 疾病发病机制及为新药研发提供帮助。因此,研究真核蛋白质的亚细胞位点意义十分重大。随着基因组测序的完成,真核蛋白质 序列信息增长迅速,为真核蛋白质亚细胞位点的研究提出了更多的挑战。传统的实验法难以满足蛋白质信息量迅速增长的需求。 而采用生物信息学手段处理大规模数据的计算预测方法,可在较短时间内获得大量真核蛋白质亚细胞位点信息,弥补了实验法 的不足。因此,运用计算预测法预测真核蛋白质的亚细胞位点成为生物信息学领域的研究热点之一。本文主要从提取真核蛋白质 的特征信息、计算预测方法及预测效果的评价三个方面,介绍近年来真核蛋白质亚细胞位点预测的研究进展。     

Discovering amino acid patterns on binding sites in protein complexes

Discovering amino acid (AA) patterns on protein binding sites has recently become popular. We propose a method to discover the association relationship among AAs on binding sites. Such knowledge of binding sites is very helpful in predicting protein-protein interactions. In this paper, we focus on protein complexes which have protein-protein recognition. The association rule mining technique is used to discover geographically adjacent amino acids on a binding site of a protein complex. When mining, instead of treating all AAs of binding sites as a transaction, we geographically partition AAs of binding sites in a protein complex. AAs in a partition are treated as a transaction. For the partition process, AAs on a binding site are projected from three-dimensional to two-dimensional. And then, assisted with a circular grid, AAs on the binding site are placed into grid cells. A circular grid has ten rings: a central ring, the second ring with 6 sectors, the third ring with 12 sectors, and later rings are added to four sectors in order. As for the radius of each ring, we examined the complexes and found that 10Å is a suitable range, which can be set by the user. After placing these recognition complexes on the circular grid, we obtain mining records (i.e. transactions) from each sector. A sector is regarded as a record. Finally, we use the association rule to mine these records for frequent AA patterns. If the support of an AA pattern is larger than the predetermined minimum support (i.e. threshold), it is called a frequent pattern. With these discovered patterns, we offer the biologists a novel point of view, which will improve the prediction accuracy of protein-protein recognition. In our experiments, we produced the AA patterns by data mining. As a result, we found that arginine (arg) most frequently appears on the binding sites of two proteins in the recognition protein complexes, while cysteine (cys) appears the fewest. In addition, if we discriminate the shape of binding sites between concave and convex further, we discover that patterns {arg, glu, asp} and {arg, ser, asp} on the concave shape of binding sites in a protein more frequently (i.e. higher probability) make contact with {lys} or {arg} on the convex shape of binding sites in another protein. Thus, we can confidently achieve a rate of at least 78%. On the other hand {val, gly, lys} on the convex surface of binding sites in proteins is more frequently in contact with {asp} on the concave site of another protein, and the confidence achieved is over 81%. Applying data mining in biology can reveal more facts that may otherwise be ignored or not easily discovered by the naked eye. Furthermore, we can discover more relationships among AAs on binding sites by appropriately rotating these residues on binding sites from a three-dimension to two-dimension perspective. We designed a circular grid to deposit the data, which total to 463 records consisting of AAs. Then we used the association rules to mine these records for discovering relationships. The proposed method in this paper provides an insight into the characteristics of binding sites for recognition complexes.     

A general method of deriving the best binding site model consistent with experimental binding data

An analysis of binding data is presented which yields the best binding site model consistent with the experimental data. The analysis is applicable to homotropic binding and yields the number of independent sites, number of interacting sites (dimers and tetramers of sites), intrinsic association constants, and degree of interaction. The information is derived from the roots of a binding polynomial constructed by the fitted Adair constants.     

Identification of thromboxane synthase amino acid residues involved in heme-propionate binding

Thromboxane A2 synthase (TXAS) is a member of the cytochrome P450 superfamily and catalyzes an isomerization reaction that converts prostaglandin H2 to thromboxane A2. As a step toward understanding the structure/function relationships of TXAS, we mutated amino acid residues predicted to bind the propionate groups of A- and D-pyrrole rings of the heme. These mutations at each of these residues (Asn-110, Trp-133, Arg-137, Arg-413, and Arg-478) resulted in altered heme binding, as evidenced by perturbation of the absorption spectra and EPR. The mutations, although causing no significant changes in the secondary structure of the proteins, induced tertiary structural changes that led to increased susceptibility to trypsin digestion and alteration of the intrinsic protein fluorescence. Moreover, these mutant proteins lost their binding affinity to the substrate analog, had a lower heme content and retained less than 5% of the wild-type catalytic activity. However, mutations at the neighboring amino acid of the aforementioned residues yielded mutant proteins retaining the biochemical and biophysical properties of the wild type TXAS. Aligning the TXAS sequence with the structurally known P450s, we proposed that in TXAS the A-ring propionate of the heme is hydrogen bonded to Asn-110, Arg-413, and Arg-478, whereas D-ring propionate is hydrogen bonded to Trp-133 and Arg-137. Furthermore, both A- and D-ring propionates bulge away from the heme plane and both lie on the proximal face of heme plane, a structure similar to P450terp.     

BiP and zein binding domains within the delta zein protein

Zeins are alcohol soluble seed storage proteins synthesized within the endosperm of maize and subsequently deposited into endoplasmic reticulum (ER) derived protein bodies. The genes encoding the beta and delta zeins were previously introduced into tobacco with the expectation of improving the nutritional quality of plants (Bagga et al. in Plant Physiol 107:13, 1997). Novel protein bodies are produced in the leaves of transgenic plants accumulating the beta or delta zein proteins. The mechanism of protein body formation within leaves is unknown. It is also unknown how zeins are retained in the ER since they do not contain known ER retention motifs. Retention may be due to an interaction of zeins with an ER chaperone such as binding luminal protein (BiP). We have demonstrated protein–protein interactions with the delta zeins, beta zeins, and BiP proteins using an E. coli two-hybrid system. In this study, four putative BiP binding motifs were identified within the delta zein protein using a BiP scoring program (Blond-Elguindi et al. in Cell 75:717, 1993). These putative binding motifs were mutated and their effects on protein interactions were analyzed in both a prokaryotic two-hybrid system and in plants. These mutations resulted in reduced BiP–zein protein interaction and also altered zein–zein interactions. Our results indicate that specific motifs are necessary for BiP–delta zein protein interactions and that there are specific motifs which are necessary for zein–zein interactions. Furthermore, our data demonstrates that zein proteins must be able to interact with BiP and zeins for their stability and ability to form protein bodies.     

Staurosporine-based binding assay for testing the affinity of compounds to protein kinases

Staurosporine is a broad-spectrum inhibitor of both tyrosine and serine/threonine protein kinases. Excitation of staurosporine and its analogues at 296 nm results in major emission bands centered at 378 and 396 nm. The intensity of the emission bands is enhanced on binding to the adenosine triphosphate (ATP) site of many protein kinases. This property was used to develop a competitive displacement assay for evaluating the binding affinity of small molecules to protein kinases. The assay was validated in both cuvette and plate formats for several phosphorylated and non-phosphorylated protein kinases. The throughput of the assay is high enough to be used in drug discovery for screening as well as lead optimization.     

The specificity and kinetic properties of trypsin ethylated at the binding site

Treatment of trypsin with triethyloxonium tetrafluoroborate at pH 8, 25 °C, results in abolition of binding to the enzyme of specific cationic substrates and inhibitors. The binding constant of soybean trypsin inhibitor to ethylated trypsin is 10000-fold smaller than to intact trypsin. However, the intrinsic ability of trypsin to recognize and react with nonspecific neutral substrates and inhibitors is not lost, and in several cases even considerably enhanced. Thus ethylated trypsin (Tret) resembles chymotrypsin in its behavior. Trypsin-like enzymes are also affected in a similar manner.     

Phylogeny reconstruction for protein sequences based on amino acid properties

This paper presents an essentially new method used to construct phylogenetic trees from related amino acid sequences. The method is based on a new distance measure which describes sequence relationships by means of typical steric and physicochemical properties of the amino acids and is advantageous in some essential points. The method was applied to different sets of protein sequences and the results were compared with other well-established methods.     

Detection of proteins based on amino acid sequences by multiple aptamers against tripeptides

A number of different ligands have been tested in the course of the development of protein array technology. The most extensively studied example of protein ligands has been based on antibody-antigen interaction. Other examples include protein-protein, protein-nucleic acid, and protein-small molecule interactions. All these ligands can recognize and specifically bind to protein epitopes. In this study, we have developed a novel technology using DNA-based aptamers to detect proteins based on their amino acid sequences. Mouse cathepsin D was used for the proof of principle experiment. Four tripeptides, Leu-Ala-Ser, Asp-Gly-Ile, Gly-Glu-Leu, and Lys-Ala-Ile, were selected based on the published amino acid sequence of mouse cathepsin D. DNA aptamers against the tripeptides were isolated using the systematic evolution of ligands of exponential enrichment method. We have demonstrated that the aptamers specifically interacted with mouse cathepsin D using the structure-switch method. We further performed a proximity-dependent ligation assay to demonstrate that multiple aptamers could specifically detect the protein from cell extracts. In principle, one library containing 8000 aptamers should be enough to detect almost all proteins in the whole proteome in all organisms. This technology could be applied to generate a new generation of protein arrays.     

Efficient docking of peptides to proteins without prior knowledge of the binding site

Reliability in docking of ligand molecules to proteins or other targets is an important challenge for molecular modeling. Applications of the docking technique include not only prediction of the binding mode of novel drugs, but also other problems like the study of protein-protein interactions. Here we present a study on the reliability of the results obtained with the popular AutoDock program. We have performed systematical studies to test the ability of AutoDock to reproduce eight different protein/ligand complexes for which the structure was known, without prior knowledge of the binding site. More specifically, we look at factors influencing the accuracy of the final structure, such as the number of torsional degrees of freedom in the ligand. We conclude that the Autodock program package is able to select the correct complexes based on the energy without prior knowledge of the binding site. We named this application blind docking, as the docking algorithm is not able to "see" the binding site but can still find it. The success of blind docking represents an important finding in the era of structural genomics.