Tree-based disease classification using protein data

A reliable and precise classification of diseases is essential for successful diagnosis and treatment. Using mass spectrometry from clinical specimens, scientists may find the protein variations among disease and use this information to improve diagnosis. In this paper, we propose a novel procedure to classify disease status based on the protein data from mass spectrometry. Our new tree-based algorithm consists of three steps: projection, selection and classification tree. The projection step aims to project all observations from specimens into the same bases so that the projected data have fixed coordinates. Thus, for each specimen, we obtain a large vector of 'coefficients' on the same basis. The purpose of the selection step is data reduction by condensing the large vector from the projection step into a much lower order of informative vector. Finally, using these reduced vectors, we apply recursive partitioning to construct an informative classification tree. This method has been successfully applied to protein data, provided by the Department of Radiology and Chemistry at Duke University.     

Improved detection of substantia nigra pathology in Alzheimer's disease.

The role of substantia nigra pathology in Alzheimer's disease (AD) is uncertain. Detection of pathology may be obscured by intraneuronal neuromelanin and influenced by stains. We determined methods for optimal visualization of nigral pathology in 45 cases of AD. For detection of Lewy bodies (LBs), we compared ubiquitin and alpha-synuclein immunostains to hematoxylin and eosin (H&E). For neurofibrillary tangles (NFTs) and neuropil threads (NTs), we compared Gallyas silver and paired helical filament (PHF) immunostains, after bleaching of melanin, to modified Bielschowsky, Gallyas, and PHF alone. The number of LB cases was not different using the three stains. However, more LBs per section were detected using alpha-synuclein (z=4.88, p<0.001). Twice the number of cases exhibited NFT (z=8.21; p<0.001) and the mean NFT number per section was 2.8-5.2-fold greater, using Gallyas and PHF after bleaching compared to without bleaching (chi(2)=142.17; p<0.001). More NTs (z=6.54; p<0.001) were observed with PHF and Gallyas after bleaching. With optimal methods, we found LBs in 27%, NFTs in 89%, and NTs in all 45 AD cases. We show that detection of nigra pathology is influenced by histological method. Clinicopathological studies using these methods are needed to determine the role of nigral pathology in AD.     

Improved data retrieval from TreeBASE via taxonomic and linguistic data enrichment

Background  

TreeBASE, the only data repository for phylogenetic studies, is not being used effectively since it does not meet the taxonomic data retrieval requirements of the systematics community. We show, through an examination of the queries performed on TreeBASE, that data retrieval using taxon names is unsatisfactory.     

Protocols for disease classification from mass spectrometry data

We report our results in classifying protein matrix-assisted laser desorption/ionization-time of flight mass spectra obtained from serum samples into diseased and healthy groups. We discuss in detail five of the steps in preprocessing the mass spectral data for biomarker discovery, as well as our criterion for choosing a small set of peaks for classifying the samples. Cross-validation studies with four selected proteins yielded misclassification rates in the 10-15% range for all the classification methods. Three of these proteins or protein fragments are down-regulated and one up-regulated in lung cancer, the disease under consideration in this data set. When cross-validation studies are performed, care must be taken to ensure that the test set does not influence the choice of the peaks used in the classification. Misclassification rates are lower when both the training and test sets are used to select the peaks used in classification versus when only the training set is used. This expectation was validated for various statistical discrimination methods when thirteen peaks were used in cross-validation studies. One particular classification method, a linear support vector machine, exhibited especially robust performance when the number of peaks was varied from four to thirteen, and when the peaks were selected from the training set alone. Experiments with the samples randomly assigned to the two classes confirmed that misclassification rates were significantly higher in such cases than those observed with the true data. This indicates that our findings are indeed significant. We found closely matching masses in a database for protein expression in lung cancer for three of the four proteins we used to classify lung cancer. Data from additional samples, increased experience with the performance of various preprocessing techniques, and affirmation of the biological roles of the proteins that help in classification, will strengthen our conclusions in the future.     

Identification of candidate IgG biomarkers for Alzheimer's disease via combinatorial library screening

The adaptive immune system is thought to be a rich source of protein biomarkers, but diagnostically useful antibodies remain unknown for a large number of diseases. This is, in part, because the antigens that trigger an immune response in many diseases remain unknown. We present here a general and unbiased approach to the identification of diagnostically useful antibodies that avoids the requirement for antigen identification. This method involves the comparative screening of combinatorial libraries of unnatural, synthetic molecules against serum samples obtained from cases and controls. Molecules that retain far more IgG antibodies from the case samples than the controls are identified and subsequently tested as capture agents for diagnostically useful antibodies. The utility of this method is demonstrated using a mouse model for multiple sclerosis and via the identification of two candidate IgG biomarkers for Alzheimer's disease.     

Regularized binormal ROC method in disease classification using microarray data

Background  

An important application of microarrays is to discover genomic biomarkers, among tens of thousands of genes assayed, for disease diagnosis and prognosis. Thus it is of interest to develop efficient statistical methods that can simultaneously identify important biomarkers from such high-throughput genomic data and construct appropriate classification rules. It is also of interest to develop methods for evaluation of classification performance and ranking of identified biomarkers.     

Improved variance estimation of classification performance via reduction of bias caused by small sample size

Background  

Supervised learning for classification of cancer employs a set of design examples to learn how to discriminate between tumors. In practice it is crucial to confirm that the classifier is robust with good generalization performance to new examples, or at least that it performs better than random guessing. A suggested alternative is to obtain a confidence interval of the error rate using repeated design and test sets selected from available examples. However, it is known that even in the ideal situation of repeated designs and tests with completely novel samples in each cycle, a small test set size leads to a large bias in the estimate of the true variance between design sets. Therefore different methods for small sample performance estimation such as a recently proposed procedure called Repeated Random Sampling (RSS) is also expected to result in heavily biased estimates, which in turn translates into biased confidence intervals. Here we explore such biases and develop a refined algorithm called Repeated Independent Design and Test (RIDT).     

Neural network on interval-censored data with application to the prediction of Alzheimer's disease

Alzheimer's disease (AD) is a progressive and polygenic disorder that affects millions of individuals each year. Given that there have been few effective treatments yet for AD, it is highly desirable to develop an accurate model to predict the full disease progression profile based on an individual's genetic characteristics for early prevention and clinical management. This work uses data composed of all four phases of the Alzheimer's Disease Neuroimaging Initiative (ADNI) study, including 1740 individuals with 8 million genetic variants. We tackle several challenges in this data, characterized by large-scale genetic data, interval-censored outcome due to intermittent assessments, and left truncation in one study phase (ADNIGO). Specifically, we first develop a semiparametric transformation model on interval-censored and left-truncated data and estimate parameters through a sieve approach. Then we propose a computationally efficient generalized score test to identify variants associated with AD progression. Next, we implement a novel neural network on interval-censored data (NN-IC) to construct a prediction model using top variants identified from the genome-wide test. Comprehensive simulation studies show that the NN-IC outperforms several existing methods in terms of prediction accuracy. Finally, we apply the NN-IC to the full ADNI data and successfully identify subgroups with differential progression risk profiles. Data used in the preparation of this article were obtained from the ADNI database.     

Gene identification and protein classification in microbial metagenomic sequence data via incremental clustering

Background  

The identification and study of proteins from metagenomic datasets can shed light on the roles and interactions of the source organisms in their communities. However, metagenomic datasets are characterized by the presence of organisms with varying GC composition, codon usage biases etc., and consequently gene identification is challenging. The vast amount of sequence data also requires faster protein family classification tools.     

Kernel-imbedded Gaussian processes for disease classification using microarray gene expression data

Background  

Designing appropriate machine learning methods for identifying genes that have a significant discriminating power for disease outcomes has become more and more important for our understanding of diseases at genomic level. Although many machine learning methods have been developed and applied to the area of microarray gene expression data analysis, the majority of them are based on linear models, which however are not necessarily appropriate for the underlying connection between the target disease and its associated explanatory genes. Linear model based methods usually also bring in false positive significant features more easily. Furthermore, linear model based algorithms often involve calculating the inverse of a matrix that is possibly singular when the number of potentially important genes is relatively large. This leads to problems of numerical instability. To overcome these limitations, a few non-linear methods have recently been introduced to the area. Many of the existing non-linear methods have a couple of critical problems, the model selection problem and the model parameter tuning problem, that remain unsolved or even untouched. In general, a unified framework that allows model parameters of both linear and non-linear models to be easily tuned is always preferred in real-world applications. Kernel-induced learning methods form a class of approaches that show promising potentials to achieve this goal.     

Fibrillar amyloid-beta peptides activate microglia via TLR2: implications for Alzheimer's disease

Microglial activation is an important pathological component in brains of patients with Alzheimer's disease (AD), and fibrillar amyloid-beta (Abeta) peptides play an important role in microglial activation in AD. However, mechanisms by which Abeta peptides induce the activation of microglia are poorly understood. The present study underlines the importance of TLR2 in mediating Abeta peptide-induced activation of microglia. Fibrillar Abeta1-42 peptides induced the expression of inducible NO synthase, proinflammatory cytokines (TNF-alpha, IL-1beta, and IL-6), and integrin markers (CD11b, CD11c, and CD68) in mouse primary microglia and BV-2 microglial cells. However, either antisense knockdown of TLR2 or functional blocking Abs against TLR2 suppressed Abeta1-42-induced expression of proinflammatory molecules and integrin markers in microglia. Abeta1-42 peptides were also unable to induce the expression of proinflammatory molecules and increase the expression of CD11b in microglia isolated from TLR2(-/-) mice. Finally, the inability of Abeta1-42 peptides to induce the expression of inducible NO synthase and to stimulate the expression of CD11b in vivo in the cortex of TLR2(-/-) mice highlights the importance of TLR2 in Abeta-induced microglial activation. In addition, ligation of TLR2 alone was also sufficient to induce microglial activation. Consistent to the importance of MyD88 in mediating the function of various TLRs, antisense knockdown of MyD88 also inhibited Abeta1-42 peptide-induced expression of proinflammatory molecules. Taken together, these studies delineate a novel role of TLR2 signaling pathway in mediating fibrillar Abeta peptide-induced activation of microglia.     

Analysis of cohort studies with multivariate and partially observed disease classification data

Complex diseases like cancers can often be classified into subtypes using various pathological and molecular traits of the disease. In this article, we develop methods for analysis of disease incidence in cohort studies incorporating data on multiple disease traits using a two-stage semiparametric Cox proportional hazards regression model that allows one to examine the heterogeneity in the effect of the covariates by the levels of the different disease traits. For inference in the presence of missing disease traits, we propose a generalization of an estimating equation approach for handling missing cause of failure in competing-risk data. We prove asymptotic unbiasedness of the estimating equation method under a general missing-at-random assumption and propose a novel influence-function-based sandwich variance estimator. The methods are illustrated using simulation studies and a real data application involving the Cancer Prevention Study II nutrition cohort.     

阿茨海默氏病研究

阿茨海默氏病（Alzheimer‘s disease,AD）受到科学界的广泛关注。已发现的AD相关基因的突变,只能解释某有族性病例,而至少60％的AD患者没有家族史,对这些散发性AD的病理,van Leeuwen等做了有意义 探索,他们的实验证明,AD脑部存在由于GA缺失造成移码突变的β淀粉样蛋白前体和泛素－B,并推测这种移码突变是AD病理的重要起妈因子。该实验开辟了从蛋白合成错误的角度研究AD的新视点,并为RNA编辑提供了新的类型。     

Causes of Alzheimer's disease

It is now understood that genetic factors play a crucial role in the risk of developing Alzheimer''s disease (AD). Rare mutations in at least 3 genes are responsible for early-onset familial AD. A common polymorphism in the apolipoprotein E gene is the major determinant of risk in families with late-onset AD, as well as in the general population. Advanced age, however, remains the major established risk factor for AD, although environmental variables may also have some role in disease expression. Some pathogenic factors directly associated with aging include oxidative damage and mutations in messenger RNA. Other factors unrelated to the aging process may, in the future, be amenable to therapeutic intervention by way of estrogen replacement therapy for postmenopausal women, anti-inflammatory drug therapy and reducing vascular risk factors. Older theories, such as aluminum playing a role in the pathogenesis of AD, have been mostly discarded as our understanding of pathogenic mechanisms of AD has advanced.