基于免疫GA与Gibbs的模体识别算法

生物序列motif识别问题是当今生物信息学面临的一个复杂问题,要设计一个能识别所有motif的方法几乎是不可能的。针对该问题,在免疫遗传算法中引入了统计估计,提高了motif识别的精度,根据个体的浓度和适应值概率。设计了免疫替换算子,有效地解决了种群的多样性问题,利用Gibbs Sampler算法生成种子,提高了免疫遗传算法的搜索速度,最后得到了一个基于免疫GA与Gibbs Sampler的生物序列motif识别算法,该算法充分发挥了免疫遗传算法和Gibbs Sampler算法的优越性,较好地解决了计算速度和计算精度之间的矛盾。实验表明,该算法是有效的。     

单体型组装MEC问题的参数化算法研究

单体型组装MEC问题指如何利用个体的DNA测序片断数据,翻转最少的SNP位点值以确定该个体单体型的计算问题。根据片段数据的特点提出了一个时间复杂度为 O（nk₂2^k₂+mlogm+mk₁）的参数化算法,其中m为片段数,n为单体型的SNP位点数,k1为一个片断覆盖的最大SNP位点数（通常小于10）,k₂为覆盖同一SNP位点的片段的最大数（通常不大于10）。对于实际DNA测序中的片段数据,即使m和n都相当大,该算法也可以在较短的时间得到MEC问题的精确解,具有良好的可扩展性和较高的实用价值。     

DNA序列拼接的分布式并行处理

针对分布式存储环境,本文提出一种DNA序列拼接的并行算法,分别对序列拼接中OVERAP、LAYOUT和CONSENSUS阶段的串行处理过程和并行算法进行了描述,并给出了算法复杂性分析。数值试验结果表明,算法是高效的。     

Multiple Recombination Events and Strong Purifying Selection at the Origin of SARS-CoV-2 Spike Glycoprotein Increased Correlated Dynamic Movements

Our evolutionary and structural analyses revealed that the severe acute respiratory syndrome (SARS) coronavirus 2 (SARS-CoV-2) spike gene is a complex mosaic resulting from several recombination events. Additionally, the fixation of variants has mainly been driven by purifying selection, suggesting the presence of conserved structural features. Our dynamic simulations identified two main long-range covariant dynamic movements of the novel glycoprotein, and showed that, as a result of the evolutionary duality, they are preserved. The first movement involves the receptor binding domain with the N-terminal domain and the C-terminal domain 2 and is maintained across human, bat and pangolin coronaviruses. The second is a complex network of long-range dynamics specific to SARS-CoV-2 involving the novel PRRA and the conserved KR*SF cleavage sites, as well as conserved segments in C-terminal domain 3. These movements, essential for host cell binding, are maintained by hinges conserved across human, bat, and pangolin coronaviruses glycoproteins. The hinges, located around Threonine 333 and Proline 527 within the N-terminal domain and C-terminal domain 2, represent candidate targets for the future development of novel pan-coronavirus inhibitors. In summary, we show that while recombination created a new configuration that increased the covariant dynamic movements of the SARS-CoV-2 glycoprotein, negative selection preserved its inter-domain structure throughout evolution in different hosts and inter-species transmissions.     

枸杞子促进虹膜色素上皮细胞定向分化治疗视网膜色素变性的分子网络调控机制研究

目的:基于生物信息学和网络药理学方法探究枸杞子（LF）促进虹膜色素上皮细胞定向分化治疗视网膜色素变性（RP）的分子网络调控机制。方法:在中药系统药理学数据库和分析平台（TCMSP）检索LF的主要活性成分及其可能的作用靶点；从GEO数据库下载GSE81058基因表达谱数据，采用R软件构建虹膜色素上皮细胞（IPE）和视网膜色素上皮细胞（RPE）差异表达基因火山图和聚类热图；采用STRING数据库构建LF成分-靶点和IPE-RPE差异表达基因的蛋白质-蛋白质交互网络（PPI），运用Cytoscape提取交集网络基因；通过DAVID数据库对交集基因进行基因本体和信号通路分析；采用cytoHubba分析LF作用于IPE的关键靶点。结果:共获得LF的化学成分188个，其中主要的入血活性成分36个，这些活性成分可能作用的靶点201个；IPE与RPE主要的差异表达基因有142个，其中上、下调基因各71个；发掘得到105个与LF促进IPE分化以治疗RP的相关基因靶点；LF作用于这些靶点涉及的生物学过程包括调节细胞分化、调节视网膜层形成等，相关的分子功能主要涉及转录共激活剂活性、离子通道活性等，主要富集于细胞质、高尔基体、细胞核等区域；LF作用于IPE涉及的分子机制包括调控PI3K-Akt信号通路、神经活性配体-受体相互作用通路等；LF促进IPE分化主要作用于SPP1、MMP9等十个关键靶点。结论:IPE与RPE具有明显差异的基因表达谱，LF的有效活性成分可作用于这些差异表达基因，促进IPE细胞分化为RPE，从而治疗RP。本研究可为细胞移植与中医药联合的中西医结合疗法应用于RP的临床治疗提供科学依据。     

Mass spectrometry analysis of nitrotyrosine‐containing proteins

Oxidative stress plays important roles in a wide range of diseases such as cancer, inflammatory disease, neurodegenerative disorders, etc. Tyrosine nitration in a protein is a chemically stable oxidative modification, and a marker of oxidative injuries. Mass spectrometry (MS) is a key technique to identify nitrotyrosine‐containing proteins and nitrotyrosine sites in endogenous and synthetic nitroproteins and nitropeptides. However, in vivo nitrotyrosine‐containing proteins occur with extreme low‐abundance to severely challenge the use of MS to identify in vivo nitroproteins and nitrotyrosine sites. A preferential enrichment of nitroproteins and/or nitropeptides is necessary before MS analysis. Current enrichment methods include immuno‐affinity techniques, chemical derivation of the nitro group plus target isolations, followed with tandem mass spectrometry analysis. This article reviews the MS techniques and pertinent before‐MS enrichment techniques for the identification of nitrotyrosine‐containing proteins. This article reviews future trends in the field of nitroproteomics, including quantitative nitroproteomics, systems biological networks of nitroproteins, and structural biology study of tyrosine nitration to completely clarify the biological functions of tyrosine nitration. © 2013 Wiley Periodicals, Inc. Mass Spec Rev 34: 423–448, 2015.     

Identification of New Potential Interaction Partners for Human Cytoplasmic Copper Chaperone Atox1: Roles in Gene Regulation?

The human copper (Cu) chaperone Atox1 delivers Cu to P_1B type ATPases in the Golgi network, for incorporation into essential Cu-dependent enzymes. Atox1 homologs are found in most organisms; it is a 68-residue ferredoxin-fold protein that binds Cu in a conserved surface-exposed Cys-X-X-Cys (CXXC) motif. In addition to its well-documented cytoplasmic chaperone function, in 2008 Atox1 was suggested to have functionality in the nucleus. To identify new interactions partners of Atox1, we performed a yeast two-hybrid screen with a large human placenta library of cDNA fragments using Atox1 as bait. Among 98 million fragments investigated, 25 proteins were found to be confident interaction partners. Nine of these were uncharacterized proteins, and the remaining 16 proteins were analyzed by bioinformatics with respect to cell localization, tissue distribution, function, sequence motifs, three-dimensional structures and interaction networks. Several of the hits were eukaryotic-specific proteins interacting with DNA or RNA implying that Atox1 may act as a modulator of gene regulation. Notably, because many of the identified proteins contain CXXC motifs, similarly to the Cu transport reactions, interactions between these and Atox1 may be mediated by Cu.