Identification of two-dimensional electrophoresis-separated proteins in human hepatoma cell by electrospray ion trap mass spectrometry

As one of the most important analytical methods in proteome research, mass spectrometry was utilized to identify proteins separated by two-dimensional electrophoresis in the human hepatoma cell line BEL-7404. The protein spots were excised from the gel, followed by in-gel digestion, and the peptide mappings were analyzed by liquid chromatography electrospray ion trap mass spectrometer. Nine proteins were identified via database searching, according to the molecular weights and amino acid sequences of peptides, among which two proteins have not been identified in the other liver-cell database. The sequence coverage was 21% –72%. Furthermore, the relationship between the expressed proteins and the liver carcinoma was discussed.     

基于随机统计方法的地下水污染源反演识别

地下水污染具有存在的隐蔽性、发现的滞后性特点,这给地下水污染修复方案设计、污染风险评估、污染责任认定都带来了很大的困难。以某垃圾填埋场垃圾渗滤液地下泄漏为例,对地下水污染源的反演识别问题进行研究。通过污染质运移模拟模型、替代模型、随机统计方法(伴随状态方法及贝叶斯方法)的综合运用,反演识别地下水污染源的位置、个数及释放历史。结果表明,伴随状态方法和贝叶斯方法的联合运用可以反演识别出污染源的个数、位置及释放历史;贝叶斯方法反演识别污染源过程中,以替代模型作为模拟模型的转化形式,用替代模型代替模拟模型大幅度地减小了计算负荷,并保持较高的精度。     

Predominant naturally processed peptides bound to HLA-DR1 are derived from MHC-related molecules and are heterogeneous in size.

Peptides bound to class I molecules are 8-10 amino acids long, and possess a binding motif representative of peptides that bind to a given class I allele. In the only published study of naturally processed peptides bound to class II molecules (mouse I-Ab and I-Eb), these peptides were longer (13-17 amino acids) and had heterogenous carboxy terminals but precise amino-terminal truncations. Here we report the characterization of acid-eluted peptides bound to HLA-DR1 by high-performance liquid chromatography, mass spectrometry and microsequencing analyses. The relative molecular masses of the peptides varied between 1,602 and 2,996 (13-25 residues), the most abundant individual M(r) values being between 1,700 and 1,800, corresponding to an average peptide length of 15 residues. Complete sequence data were obtained for twenty peptides derived from five epitopes, of which all but one were from self proteins. These peptides represented sets nested at both the N- and C-terminal ends. Binding experiments confirmed that all of the isolated peptides had high affinity for the groove of DR1. Alignment of the peptides bound to HLA-DR1 and the sequences of 35 known HLA-DR1-binding peptides revealed a putative motif. Although peptides bound to class II molecules may have some related features (due to the nonpolymorphic HLA-DR alpha-chain), accounting for degenerate binding to different alleles, particular amino acids in the HLA-DR beta-chains presumably define allelic specificity of peptide binding.     

Technical development of proteomics

Proteome research is an important complementarity for the genome research, which uses quantitative protein-level measurements to characterize biological process. Proteome techniques include two aspects. One is two-dimensional electrophoresis, which is the unique technology to separate the components of proteome. The other is identification of the separated spots, which is the mainstay of the proteome. Image analysis can define the primary attributes, such as isoelectric point and molecular weight; microsequencing is the elementary tool; peptide mass fingerprinting and peptide fragment sequencing via mass spectrometry can get higher resolution and accuracy; amino acid composition analysis can also be used to identify the proteins. Application of these techniques in proteomics depends on various database, software and network as well as high throughput screening.     

Global landscape of protein complexes in the yeast Saccharomyces cerevisiae

Identification of protein-protein interactions often provides insight into protein function, and many cellular processes are performed by stable protein complexes. We used tandem affinity purification to process 4,562 different tagged proteins of the yeast Saccharomyces cerevisiae. Each preparation was analysed by both matrix-assisted laser desorption/ionization-time of flight mass spectrometry and liquid chromatography tandem mass spectrometry to increase coverage and accuracy. Machine learning was used to integrate the mass spectrometry scores and assign probabilities to the protein-protein interactions. Among 4,087 different proteins identified with high confidence by mass spectrometry from 2,357 successful purifications, our core data set (median precision of 0.69) comprises 7,123 protein-protein interactions involving 2,708 proteins. A Markov clustering algorithm organized these interactions into 547 protein complexes averaging 4.9 subunits per complex, about half of them absent from the MIPS database, as well as 429 additional interactions between pairs of complexes. The data (all of which are available online) will help future studies on individual proteins as well as functional genomics and systems biology.     

多肽序列优化降低c-Src蛋白靶向药物的毒性风险

避免c-Src蛋白的多肽类拮抗剂与多个蛋白发生混杂性结合,对于降低抗癌药物的毒性风险具有重要作用。本研究运用生物信息学方法对氨基酸序列进行优化设计,旨在减少混杂性结合的发生。本研究综合利用各种多肽数据库和生物信息学工具,首先总结了多肽分子与多个蛋白SH3结构域之间潜在的混杂性结合,并发现了其中的内在规律。随后,根据所发现的规律,对多肽的氨基酸序列进行针对性的优化设计。结果表明,大多数多肽在经过优化后,所结合的c-Src以外的蛋白数量都有所下降,从而显著提高了多肽与c-Src蛋白之间结合的特异性(P0.05),并降低了其潜在的毒性风险。本研究所取得的结果将为设计具有高度特异性和低毒性的靶向性多肽药物提供参考。     

复杂环境下基于动态贝叶斯网络的目标识别

为了提升高动态复杂电磁环境下空战过程中对目标的识别能力,针对SBN网络模型无法满足战场的动态性要求以及对目标的经常性误识别问题,设计了一种基于变结构动态贝叶斯网络的目标类型识别模型。该模型是由静态贝叶斯网络模型演变而来,具有良好的动态表达性和滤波功能,弥补了SBN的不足,并且对空战过程中目标特征信息丢失的问题有良好的容错能力。仿真结果表明,基于动态贝叶斯网络的目标识别的识别效果,优于基于参数学习贝叶斯网络的目标识别。使用该模型后目标识别的准确性提高了5%,有效地解决目标类型识别过程中数据缺失和信息不足的问题。     

Sequence analysis of peptides with biological activities using electrospray-Fourier trans form ion cyclotron resonance mass spectrometry

The mass spectra of five peptides with biological activities are reported. All mass spectra were recorded using a 4.7-T Fourier transform ion cyclotron resonance mass spectrometer equipped with an external electrospray source. The accurate molecular weights for the five peptides prepared by solid phase synthesis were measured as 1765.9013, 1063.5420, 1092.5254, 820.3804 and 1078.5193, respectively. All the data were obtained with the external calibration. Differences between observed and theoretical monoisotopic molecular weights were in the (0.2—1.0)×10^-6 range. The complete primary sequence for the five polypeptides were determined using the method of in-source electrospray ionization/collision induced dissociation (ESI/CID). All the intact y series ions and b series ions were obtained from various peptides respectively, thus determining the sequences of the five polypeptides. We found that the measured accurate molecular mass of sample 4 was not in agreement with that expected from the planned synthetic peptide. The sequences of sample 4 were determined through analysis. The corresponding accurate masses of b series ions and y series ions were gained, which proved that it was correct to re-determine the sequences.     

Identification of specific binding proteins for a nuclear location sequence

The nuclear envelope is a selective barrier against the movement of macromolecules between the nucleus and cytoplasm. Nuclear proteins larger than relative molecular mass 20,000-40,000 are probably actively transported across the envelope through the nuclear pore complex and are directed by specific nuclear location sequences (NLS) in the proteins. NLS mediate the nuclear import of isolated nuclear proteins after microinjection into whole cells and the nuclear accumulation of chimaeric proteins or of non-nuclear proteins conjugated to synthetic peptides. The best-characterized NLS is the simian virus 40 large T-antigen sequence. We have identified two proteins of rat liver by chemical cross-linking that interact with a synthetic peptide containing this sequence: this interaction is specific for a functional NLS, is saturable, and high affinity. The binding proteins are present in a post-mitochondrial supernatant, in nuclei and in a nuclear envelope fraction, which is consistent with a role in the transport of nuclear proteins from the cytoplasm to the nucleus.     

基于神经网络的多功能雷达行为辨识方法

针对多功能雷达行为状态复杂多变、难以识别的问题,构建了多功能雷达行为数据集,提出了一种基于神经网络的雷达行为辨识方法。首先对数据进行预处理,提取多功能雷达的参数特征与行为状态特征,并建立两者间的映射关系。然后通过基于贝叶斯准则的变化点检测算法对原始雷达信号脉冲序列进行分割,补齐有缺失的特征参数,构造完整的可用于训练的脉冲数组样本。最后通过数据推理丰富数据库,为数据驱动的智能识别方法提供可靠的数据准备,增强神经网络的泛化能力。针对处理后的雷达行为数据集的特点,设计BP神经网络进行训练与测试。仿真实验结果表明:训练完成的网络模型在识别过程中一定程度上克服了噪声变量等干扰的影响,正确率可以达到89%。     

基于SBN模型的Internet应用协议识别方法

针对网络流量协议标注比较困难的问题,提出一种基于贝叶斯网络的半监督学习模型,以提高Inter-net协议的识别精度.该模型首先使用少量的标注样本训练贝叶斯网络分类模型,并对未标注样本进行初始分类,然后从未标注样本中挑选分类损失最小的样本加入到训练集中并重复训练分类模型,经过多次循环训练出最终的分类器.该模型可以使用未标注样本和标注样本共同训练分类模型,非常适合于标注比较困难的Internet应用协议的识别.实验结果表明：在标注样本较少的情况下,该模型的识别精度和稳定性均优于朴素贝叶斯模型和贝叶斯网络模型,对于提高Internet协议的识别精度是有效的.     

MHC class II region encoding proteins related to the multidrug resistance family of transmembrane transporters

The T-cell immune response is directed against antigenic peptide fragments generated in intracellular compartments, the cytosol or the endocytic system. Peptides derived from cytosolic proteins, usually of biosynthetic origin, are presented efficiently to T-cell receptors by major histocompatibility complex (MHC) class I molecules, with which they assemble, probably in the endoplasmic reticulum (ER). In the absence of recognizable N-terminal signal sequences, such cytosolic peptides must be translocated across the ER membrane by a novel mechanism. Genes apparently involved in the normal assembly and transport of class I molecules may themselves be encoded in the MHC. Here we show that one of these, the rat cim gene, maps to a highly polymorphic part of the MHC class II region encoding two novel members of the family of transmembrane transporters related to multidrug resistance. Other members of this family of transporter proteins are known to be capable of transporting proteins and peptides across membranes independently of the classical secretory pathway. Such molecules are credible candidates for peptide pumps that move fragments of antigenic proteins from the cytosol into the ER.     

两种胰蛋白酶酶解羊肌肉总蛋白样品的质谱分析

摘要: 目的比较国产和进口两种胰蛋白酶对蛋白质的酶解效果,为后续蛋白质组学稳定研究提供依据。方法用超声破碎法提取羊肌肉组织总蛋白,经两种胰蛋白酶酶解后液相色谱串联质谱检测酶解肽段,将结果上传蛋白质序列数据库,利用质谱分析鉴定方法分析比较两种胰蛋白酶的酶解效果,并借助生物信息学工具对所得数据进行分析。结果两种胰蛋白酶酶解同种蛋白质样品产生的肽段长度86%以上均为6-18 个氨基酸,适用于质谱检测。酶解后肽段产生的质量国产胰蛋白酶优于进口胰蛋白酶,国产胰蛋白酶比进口胰酶酶解可多鉴定到19% 的蛋白质和34%的肽段。结论进口、国产的胰蛋白酶在酶解蛋白质方面均可用于质谱检测前蛋白质的酶解消化。但国产胰蛋白酶价格明显低于进口胰酶,更具应用优势。本研究为评价胰蛋白酶酶解效果提供了一种有效、全面的研究方法。     

基于LC-MS/MS策略建立十字花科黑腐病菌蛋白质表达谱的方法研究

【目的】建立十字花科黑腐病菌(Xanthomonas campestris pv.campestris,Xcc)的蛋白质组学研究平台,用于分离、鉴定该菌的致病相关蛋白质。【方法】Xcc 8004经过液体培养,分别提取胞内蛋白质和胞外蛋白质,对所有蛋白质进行液体酶解,通过强阳离子交换色谱预分离多肽混合物,预分离后的馏分采用EASY-nLC结合LTQ-Orbitrap质谱鉴定蛋白质表达谱。【结果】建立了基于bottom-up策略的蛋白质组鉴定方法,采用第一维离子交换预分离和第二维反相色谱分离结合,实现了高通量鉴定蛋白质组表达谱的技术体系。通过SEQUEST检索,在胞内蛋白质样品中共鉴定蛋白质数目为1595个,胞外蛋白质样品共鉴定蛋白质数目为1241个。【结论】建立的LC-MS/MS方法适合用于十字花科黑腐病菌蛋白质组学的研究,为研究微生物植物相互作用奠定了方法与理论基础。     

基于高斯混合模型的敞开式质谱重叠峰解析方法

针对敞开式质谱重叠峰影响待测物特征峰识别的问题,提出了基于高斯混合模型的重叠峰解析方法.根据手肘法和质谱图中各参数意义有效选取模型初始参数,然后研究了该方法在重叠峰不同幅值比、分离度和噪声下的解析效果,并对敞开式质谱实测冰毒碎片离子的重叠峰进行解析.结果 表明,模拟数据的各个参数在解析后变化较小,均值和标准偏差的最大相对误差分别为0.4％和2％;实测数据在解析后可有效识别目标谱峰并使信噪比最高可提升10.2％,该方法解析效果好并且具有抗噪声干扰能力,可用于敞开式质谱重叠峰的解析.     

两种用于液质联用分析的蛋白质酶解处理效果比较

【目的】分析和评估两种蛋白质溶液酶解方法,为蛋白质组学研究中的大规模质谱鉴定分析建立一种稳定的样品制备技术。【方法】分别使用TEAB法(实验组A)和TFE法(实验组B)对牛血清白蛋白(BSA)进行还原、烷基化和溶液酶解,再用LTQ-Orbitrap高分辨率质谱仪采集BSA酶切肽段信息,比较两种方法得到的数据信息,分析肽段数目、可变修饰位点和蛋白质覆盖率。【结果】实验组A(TEAB法)获得36个肽段、39个修饰位点,蛋白质平均覆盖率为59.31%;实验组B(TFE法)获得27个肽段、27个修饰位点,蛋白质平均覆盖率为47.83%,实验组A的数据指标与实验组B差异显著。【结论】采用TEAB法酶切体系能实现蛋白质高效率酶切,达到获得较高肽段覆盖率的目的,可为蛋白质组学后期的质谱分析鉴定奠定实验基础。     

转向架系统簧上质量的物理参数识别

提出了基于质量感应法、状态空间理论和模态空间理论的3种转向架系统簧上质量物理参数识别方法,设计了用于物理参数识别的定向激励测试工况,并通过仿真试验对以上参数识别方法进行验证及对比分析.结果表明,当采用质量感应法、状态空间理论和模态空间理论时,附加质量大小分别选取为系统质量的8%、17%和7%时识别精度较高;总体来看,模态空间理论的识别精度最高,状态空间理论次之,质量感应法最低;就质量参数、惯量参数、一系垂向刚度及阻尼参数识别而言,可以不考虑转向架簧上质量质心与一系悬挂下作用点位置高度差的影响;在高度差较小时,也可忽略其对结构参数识别精度的影响.     

计算机算法结合实验技术进行T细胞表位筛选

计算机算法与实验技术相结合的方法,已广泛用于各种免疫相关抗原T细胞表位的鉴定,对现在可获得的计算机预测方法,以及计算机算法与实验相结合鉴定T细胞表位的方法进行了综述,主要内容包括表位预测及抗原加工处理的预测。     

同位素标记结合中性丢失扫描鉴定白蛋白早期糖基化修饰肽段

采用 [13C6] 同位素标记的葡萄糖作为标签,在体外建立非酶糖基化反应模型,并采用MALDI-TOF对模型中的糖基化人血清白蛋白(HSA)进行验证. 以高分辨率的液质联用四级杆飞行时间质谱(HPLC/Q-TOF MS)为分析平台,经过一级质谱(LC/MS)筛选糖基化肽段. 采用target MS/MS模式对筛选出的34条肽段进行分析. 运用蛋白质组学数据分析工具pFind软件对肽段的中性丢失扫描及二级质谱(LC/MS/MS)数据进行分析,最终获得16条早期糖基化修饰肽段的序列及修饰位点信息.