基于多特征融合的蛋白质折叠子预测

蛋白质折叠子预测为启发式搜索蛋白质三级结构提供了有用的信息.目前已知的折叠子预测方法大多数基于单种特征或多种特征的简单组合,本文采用一种多特征融合方法,从蛋白质的一级序列出发,对27类折叠子进行预测.使用支持向量机作为分类器,采用多对多的多类分类策略,以氨基酸组成成分、极性、极化性、范德瓦尔斯量、疏水性和预测的二级结构作为样本的六种特征,进行多特征融合,独立样本预测总精度为59.22%,与Ding等人的结果比较提高了3.2%,结果表明多特征融合方法是一种有效的蛋白质折叠子预测方法.     

基于MATLAB的隐马尔可夫模型预测蛋白质结构类

目的 准确预测蛋白质结构类,为研究其空间结构及生物功能打下基础.方法 应用隐马尔可夫模型(HMM)预测蛋白质结构类,分别构建3-状态HMM和8-状态HMM.数据来源于Chou和Zhou构建的蛋白质数据集,分别包含有204条蛋白质序列和498条蛋白质序列,通过留一法预测其准确率.结果 所构建的3-状态HMM和8-状态HMM对全α类的预测准确率最高,尤其是3-状态HMM的预测准确率达到95％以上.与Chou数据集相比,Zhou数据集对于全β类和α/β类的预测准确率也有所提高,同时,总体预测率也提高了2％左右;但α+β类的预测准确率有所下降.结论 将整条蛋白质序列作为预测模型的输入信息所构建的HMM模型能有效地预测蛋白质的结构类.     

基于二叉树支持向量机的蛋白质结构类预测

提出了一种基于二叉树支持向量机(BT-SVM)的蛋白质结构类多类预测新方法.采用26维的向量来表示蛋白质序列的特征.BT-SVM多类分类方法能消除SVM在多分类问题中存在的不可分数据问题.采用两个经典数据集作为测试数据,通过自身一致性和n折叠交叉验证方法测试了新方法的性能.预测结果表明新方法具有良好的预测能力,与使用同一数据集的已有结果相比较,新方法的Jackknife结果和目前最好的方法取得的结果相当,可作为蛋白质结构类预测的一个工具.     

基于模糊支持向量机的膜蛋白分类研究

以氨基酸组成为特征对膜蛋白的分类,忽略了序列残基之间的相关性信息,而采用传统支持向量机算法作为分类算法,在解决多类问题时会出现分类盲区问题。针对这两种情况,计算蛋白质序列的氨基酸组成、二肽组成以及6种氨基酸相关系数,将三类特征结合,作为膜蛋白序列的特征向量;同时采用模糊支持向量机作为分类器,解决了传统支持向量机在多类数据识别中的盲区问题。测试结果表明,在相同特征输入下,模糊支持向量机分类性能优于传统支持向量机;在相同分类器的情况下,氨基酸组成、二肽组成和相关系数组合的特征选择方法的分类性能优于只使用其中一类或两类特征的方法;而采取组合特征和模糊支持向量机相结合的分类策略,在独立性数据集测试中的整体预测精度达到97%,优于现有的多种分类策略,是目前最有效的膜蛋白分类方法之一。     

运用近邻传播聚类分析进行SELDI-TOF蛋白质谱特征选择

针对如何有效分析高通量SELDI-TOF质谱数据以及筛选与肿瘤相关的蛋白质位点,提出一种基于近邻传播聚类分析的特征选择方法.首先利用t-test对SELDI数据进行初筛,然后利用近邻传播聚类分析以及零空间LDA对数据进行降维和去相关处理,最后采用SVM-RFE进行特征选择,筛选出与肿瘤判别相关的蛋白质位点.利用SVM、KNN、NB及J4.8等4个分类器,估算算法的分类性能.结果表明,在卵巢癌公共数据集OC-WCX2a和OC-WCX2b以及浙江省肿瘤医院乳腺癌数据集BC-WCX2a上显示该算法,在上述3个数据集中分类率分别达到96.43％、99.66％、90.88％,敏感性分别达到97.00％、100％、96.17％,特异性分别达到95.85％、99.08％、81.92％,并分别挑选出与肿瘤判别相关的10个蛋白位点.所提出的算法能够获得较好的分类率,有效提取出具有较好判别效果的蛋白质谱位点,有助于癌症的辅助诊断.     

基于双视角和多分类器信息融合的乳腺钼靶图像肿块分类研究

乳腺肿块良恶性分类是计算机辅助诊断(CAD)的重要环节,如何提高分类的正确率和稳定性是分类研究的重点.本研究提出了4种基于双视角和多分类器信息融合的乳腺钼靶图像肿块分类模式.其中,模式1是单视角下的多分类器融合;模式2是分别先对每个分类器在两个视角下的输出进行视角融合,再对其融合结果进行多分类器融合;模式3是分别先在每...     

基于RoughSet的特征集融合PET/CT肺部肿瘤CAD模型

针对肺部肿瘤PET/CT感兴趣区域(ROI)在高维特征表示下存在着特征相关和维数灾难问题,提出了一种基于粗糙集特征集融合的PET/CT肺部肿瘤CAD模型。首先提取肺部肿瘤ROI的8维形状特征、7维灰度特征、3维Tamura纹理特征、56维GLCM特征和24维频域特征,得到98维特征矢量;然后基于遗传算法的知识约简方法和基于属性重要度的启发式算法对提取的特征集合分别进行特征级融合得到特征子集G1、G2、G3,A1、A2、A3,降低特征矢量的维数;再次利用网格寻优算法优化核函数的SVM作为分类器分别进行融合前和融合后的分类识别比较,基于遗传算法的特征集融合和基于属性重要度的特征集融合的分类识别比较2组实验;最后以2 000幅肺部肿瘤的PET/CT图像为原始数据,采用基于粗糙集特征集融合的肺部肿瘤PET/CT计算机辅助诊断模型对肺部肿瘤进行辅助诊断。实验结果表明,经过粗糙集特征集融合的肺部肿瘤诊断识别方法能有效提高肺部肿瘤的诊断正确率,一定程度上降低了特征之间的相关性。     

基于加权模糊k近邻方法的蛋白质亚细胞位点预测

蛋白质只有在特定的亚细胞位点（如细胞核、线粒体、细胞质等）才能参与正常的生命活动，因此蛋白质的哑细胞定位信息对于了解其功能有重要的意义。提出一种应用于蛋白亚细胞定位的多模糊k近邻加权投票算法。使用PSI-BLAST搜索得到的PSSM矩阵，以及1～7阶氨基酸对的信息作为输入特征，分别建立了8个模糊k近邻分类器，最后对所有分类器的结果使用加权投票得到最终预测结果。对包含四类亚细胞位置的RH-2427数据集进行jacknife测试，总预测精度达到88．1％，好于包括单一模糊k近邻在内的多种其它预测方法。同时，该方法可以方便地扩展到对包含叶绿体、高尔基体、溶酶体等更多类亚细胞位点的预测。     

基于支持向量回归方法的蛋白残基可溶性预测

介绍了一种从蛋白质序列预测残基相对可溶性的新方法。该方法基于支持向量回归，并将序列局部信息作为输入。不同于先前的大部分预测方法仅对特定的蛋白残基相对可溶性进行状态分类，该方法预测了相对可溶性的连续值，从而比状态分类保留了蛋白质三维结构的更多信息。本研究对RS-126，Manesh-215和CB-513三个数据集进行了测试。通过比较不同的参数及窗宽模型来获得最佳结果，采用平均绝对误差、相关系数等参数来衡量预测效果，同时与多层反馈神经网络方法（RVP-Net）的实验结果比较，在3-fold情况下三个数据集预测结果的平均绝对误差均有降低，相关系数均有提高。另外，该算法采用了多序列比对作为输入，效果比单序列有所提高。采用该方法，对CB-513数据集平均绝对误差可以达到16．8％、相关系数为0．562，而用RVP-Net方法分别为18．8％和0．480。这些结论表明支持向量回归方法是蛋白质序列分析的一种有效工具。     

蛋白质超二级结构序列片段的统计分析

从蛋白质的氨基酸序列出发，用统计分析的方法，对序列相似性小于40％、分辨率低于2．57的4729个蛋白质中的超二级结构序列片断进行了统计分析，统计结果得到区分四类超二级结构的有用信息，这些统计结果对蛋白质超二级预测是非常有益的。     

Renal dysplasia and asplenia in two sibs

A family is reported in which two sibs, one male and the other female, both died within 24 hours of birth with enlarged polycystic kidneys. Postmortem histology in the second child showed gross renal dysplasia. In both children the pancreas was enlarged, nodular and cystic but the liver appeared macroscopically normal. In the second child, histological examination confirmed pancreatic fibrosis with cystic dilation of ducts, but showed portal fibrosis with bile duct proliferation in the liver.
This combination of findings is very reminiscent of those in a girl and her brother reported by Ivemark et al. (1959). The children reported here also showed absence or hypoplasia of the spleen, cardiac anomalies and other features of the Ivemark syndrome (Ivemark 1955), a quite different, usually sporadic, congenital disorder. It is suggested that the children described here have a distinct lethal congenital disorder, probably inherited in an autosomal recessive manner.     

Schizophrenia in a North Swedish geographical isolate, 1900–1977. Epidemiology, genetics and biochemistry

Over 200 schizophrenic patients belonging to three major and interrelated pedigree complexes have been investigated over the past 30 years in a North Swedish geographically isolated population, presently numbering about 6,000. An intensive investigation of a number of biochemical correlates and genetic markers in a few selected families belonging to one of the major pedigrees has indicated new strategies for the current research program.
Schizophrenia, as defined operationally, is significantly associated with decreased activities of two enzymes (1) blood platelet monoamine oxidase, (2) plasma dopamine-β-hydroxylase, and (3) with the genetic marker Gc² (group specific antigen). Both enzymes are subject to genetic variation. A positive score for linkage between schizophrenia and low plasma DBH activity has been calculated, but, so far, available data are insufficient for discrimination between linkage and partial contribution of genetically controlled low plasma DBH to the pathogenesis of the disease. Alternatively, both mechanisms could be involved.
As a model for continued research, schizophrenia is explained as based on a double dominant-recessive genotype (Aabb), representing a vulnerability which in about 50 % of cases develops into clinical schizophrenia. It is suggested that the dominant mutation (A) operates on or affects MAO activity, and that the recessive genotype (bb) is instrumental in low variates of DBH activity and very likely such variates within the normal range of physiological variation. Moreover, it is suggested that the combined effects of MAO- and DBH-reduced efficiency on the metabolism of e.g. dopamine could be an essential pathogenic mechanism for the schizophrenic illness which is segregating in this population.     

The dominant diseases of modernized societies as omega-3 essential fatty acid deficiency syndrome: Substrate beriberi

About 1900, modern food selection and processing caused widespread $\text{epidemics}$ of the B vitamin deficiency diseases of beriberi and pellagra which, for genetic reasons, often expressed as different diseases ranging from bowel and heart disease to dermatoses and psychoses. But the B vitamins merely help convert essential fatty acids (EFA) into the prostaglandin (PG) tissue regulators and it now turns out that, through hydrogenation, milling and selection of w3-poor southern foods, we have also been systematically depleting, by as much as 90%, a newly discovered trace Nordic EFA (w3) of special importance to primates and sole precursor of the PG3(4) series, even as a concurrent fiber deficiency increases body demand for EFA. Since substrate EFA is processed by many B vitamin catalysts, an EFA deficiency will mimic a $\text{panh}$ypovitaminosis B, i.e., a mixture of $\text{substrate}$ beriberi and $\text{substrate}$ pellagra resembling vitamin beriberi and pellagra but exhibiting as even more diverse $\text{endemic}$ disease. This would consitute a second stage of the Modern Malnutrition and explain why some workers now hold the dominant diseases of modermized societies to be new, nutritionally based, pellagraform yet lipid-related and to range, once again, from heart disease to psychosis. It is an assumption that our dominant diseases are unrelated to each other or are merely revealed by our diagnostic acumen and therapeutic success; and that hydrogenating millions of tons of food oils annually, to destroy the rancidity producing w3-EFA, is safe for $\text{primates}$. Extensive beriberiform disease is reported here in 32 typical cases taken from medical practice which responds strikingly to linseed oil supplements (60% w3-EFA) in confirmation of identical results in Capuchins.     

What will studies of Fulani individuals naturally exposed to malaria teach us about protective immunity to malaria?

There are an estimated over 200 million yearly cases of malaria worldwide. Despite concerted international effort to combat the disease, it still causes approximately half a million deaths every year, the majority of which are young children with Plasmodium falciparum infection in sub-Saharan Africa. Successes are largely attributed to malaria prevention strategies, such as insecticide-treated mosquito nets and indoor spraying, as well as improved access to existing treatments. One important hurdle to new approaches for the treatment and prevention of malaria is our limited understanding of the biology of Plasmodium infection and its complex interaction with the immune system of its human host. Therefore, the elimination of malaria in Africa not only relies on existing tools to reduce malaria burden, but also requires fundamental research to develop innovative approaches. Here, we summarize our discoveries from investigations of ethnic groups of West Africa who have different susceptibility to malaria.     

'Very sore nights and days': the child's experience of illness in early modern England, c.1580-1720

Sick children were ubiquitous in early modern England, and yet they have received very little attention from historians. Taking the elusive perspective of the child, this article explores the physical, emotional, and spiritual experience of illness in England between approximately 1580 and 1720. What was it like being ill and suffering pain? How did the young respond emotionally to the anticipation of death? It is argued that children’s experiences were characterised by profound ambivalence: illness could be terrifying and distressing, but also a source of emotional and spiritual fulfilment and joy. This interpretation challenges the common assumption amongst medical historians that the experiences of early modern patients were utterly miserable. It also sheds light on children’s emotional feelings for their parents, a subject often overlooked in the historiography of childhood. The primary sources used in this article include diaries, autobiographies, letters, the biographies of pious children, printed possession cases, doctors’ casebooks, and theological treatises concerning the afterlife.     

Guidelines for the Validation and Use of Immunoassays for Determination of Introduced Proteins in Biotechnology Enhanced Crops and Derived Food Ingredients

Recent advancements in agricultural biotechnology have created a need for analytical techniques to determine introduced proteins in crops enhanced through modern biotechnology techniques. These proteins are expressed in plant tissues and may be present in food ingredients. Immunoassays are ideally suited for protein detection and may be used as both quantitative and threshold methods. Microplate ELISA and lateral flow devices are two of the most commonly used immunoassay formats for agricultural biotechnology applications. This paper provides general background information and a discussion of criteria for the validation and application of immunochemical methods to the analysis of proteins introduced into plants and food ingredients using biotechnology methods. It is the result of a collaborative effort of members of the Analytical Environmental Immunochemical Consortium. This collaborative effort represents the combined expertise of several organizations to reach consensus on establishing guidelines for the validation and use of immunoassays. Further, the paper offers developers and users a consistent approach to adopting the technology as well as aid in producing accurate and meaningful results.     

Ultracryotomy: development and application (with examples from the yeast cytology) (author's transl)

The preparation steps usually necessary for obtaining ultrathin frozen sections of biological material (chemical prefixation, enclosing, cryoprotective treatment, freezing, sectioning, and post-staining the sections for transmission electron microscopy) are submitted to a critical analysis. The application of cryo-ultramicrotomy, in particularly for cytochemical purposes, is reviewed. Fundamental considerations of chemical prefixation and poststaining are supported by examples from yeast cytology. Furthermore, the efficiency of the cryo-ultramicrotomy (electron optical resolution of ultrastructural details) is demonstrated on yeast cells and protoplasts.     

HLA in North Colombia Chimila Amerindians

HLA-A,-B,-C,-DRB1 and -DQB1 alleles have been studied in Chimila Amerindians from Sabana de San Angel (North Colombian Coast) by using high resolution molecular typing. A frequent extended haplotype was found:HLA-A*24:02-B*51:10-C*15:02-BRB1*04:07-DQB1*03:02 (28.7%) which has also been described in Amerinndian Mayos Mexican population (Mexico, California Gulf, Pacific Ocean). Other haplotypes had already been found in Amerindians from Mexico (Pacific and Atlantic Coast), Peru (highlands and Amazon Basin), Bolivia and North USA. A geographic pattern according to HLA allele or haplotype frequencies is lacking in Amerindians, as already known. Also, five new extended haplotypes were found in Chimila Amerindians. Their HLA-A*24:02 high frequencies characteristic is shared with aboriginal populations of Taiwan; also, HLA-C*01:02 high frequencies are found in New Zealand Maoris, New Caledonians and Kimberly Aborigines from Australia. Finally, this study may show a model of evolutionary factors acting and rising one HLA allele frequency (-A*24:02), but not in others that belong to the same or different HLA loci.     

The universal muscular chamber. A basic unit of the genitourinary, cardiovascular, gastro-intestinal, skeletal, cutaneous, respiratory and other systems, endocameral hypertension; and spasm, the key to their idiopathic diseases and arthropathies

Starting with the integument, we see many organs are contractile sacs or multiples thereof, which tubes or bags constitute the major part of the entire body. Recognition of this basic unit and its characteristics sheds new light, individually and collectively, on many disorders previously considered unrelated. Muscular tears and perforations develop in the walls of these chambers, being no way peculiar to those organs, wherein, hydrochloric acid occurs. So, it is not necessary to explain the absence of excessive acid from patients who exhibit holes in the gastric, uterine, aortic, duodenal, rectal, pulmonary, retina, and other walls. Muscle, not acid is the great common factor relating idiopathic disorders in the gastrointestinal tract to each other and to similar diseases in other systems. When the units are linked together, the lesions tend to appear as arthropathies, i.e. at the joints. Rephrasing common-place observations, frees us from conventional, conceptual cul-de-sacs. An observation is only as good as its interpretation, so all possibilities must be considered, otherwise, we will remain blinded by our misconceptions.