Comparison between principal component analysis and independent component analysis in electroencephalograms modelling

Principal Component Analysis (PCA) is a classical technique in statistical data analysis, feature extraction and data reduction, aiming at explaining observed signals as a linear combination of orthogonal principal components. Independent Component Analysis (ICA) is a technique of array processing and data analysis, aiming at recovering unobserved signals or 'sources' from observed mixtures, exploiting only the assumption of mutual independence between the signals. The separation of the sources by ICA has great potential in applications such as the separation of sound signals (like voices mixed in simultaneous multiple records, for example), in telecommunication or in the treatment of medical signals. However, ICA is not yet often used by statisticians. In this paper, we shall present ICA in a statistical framework and compare this method with PCA for electroencephalograms (EEG) analysis.We shall see that ICA provides a more useful data representation than PCA, for instance, for the representation of a particular characteristic of the EEG named event-related potential (ERP).     

Characterization of epitopes of human secretory component on free secretory component, secretory IgA, and membrane-associated secretory component

We have compared the epitopes present in various forms of human secretory component by using a panel of hybridoma-derived antibodies elicited by immunizing mice with free secretory component (FSC) or secretory IgA (sIgA). Enzyme-linked immunosorbent binding assays (ELISA) were used to assess antibody binding to FSC- and SC-containing antigens, including sIgA isolated from milk, reduced and alkylated sIgA, and sIgA assembled in vitro by incubating dimeric IgA with FSC. Immunofluorescence assays were also used to assess binding to a human epithelial tumor cell line (HT29) that expresses secretory component as an integral protein of the plasma membrane. The results can be summarized as follows. 1) Most antibodies from fusions in which sIgA was the immunizing antigen bound preferentially to sIgA. 2) Most antibodies from fusions in which FSC was the immunizing antigen bound preferentially to FSC. 3) Antibodies that bound preferentially to sIgA invariably bound sIgA assembled in vitro; antibodies that bound preferentially to FSC invariably did not. 4) Antibodies that bound readily to both sIgA and FSC were rare in all fusions. 5) The monoclonal antibodies defined at least six classes of epitopes on SC, including epitopes that were a) FSC specific and reduction sensitive, b) FSC specific and reduction insensitive, c) sIgA specific and reduction-sensitive, d) sIgA specific and reduction insensitive, e) shared by FSC and sIgA and reduction-sensitive, and f) shared by FSC and sIgA and reduction-insensitive. 6) Antibodies that mediated intense immunofluorescent staining of secretory component on HT29 cell membranes were rare and constituted a distinct subset of those which recognized epitopes shared by FSC, reduced and alkylated sIgA, and some preparations of native sIgA. Results of these antibody-binding studies indicate that most SC epitopes are not shared by FSC and sIgA. Most SC-related epitopes on sIgA appear to be generated by the physical interaction of SC with dimeric IgA, whereas most epitopes on FSC are masked or altered by this interaction. Finally, epitopes that are shared by membrane SC and FSC and/or sIgA represent a minor and immunochemically distinct subset of epitopes on SC. The high proportion of unique epitopes on the different physical forms of SC suggest that the epitopes of this molecule are highly sensitive to its molecular environment. The monoclonal reagents described here will be useful in studying the structure and function of SC; quantitating FSC, sIgA, and membrane SC; purifying various molecular forms of SC by immunoaffinity chromatography; and localizing SC in human tissues and cultured cells by immunocytochemical techniques.     

Two component systems: physiological effect of a third component

Signal transduction systems mediate the response and adaptation of organisms to environmental changes. In prokaryotes, this signal transduction is often done through Two Component Systems (TCS). These TCS are phosphotransfer protein cascades, and in their prototypical form they are composed by a kinase that senses the environmental signals (SK) and by a response regulator (RR) that regulates the cellular response. This basic motif can be modified by the addition of a third protein that interacts either with the SK or the RR in a way that could change the dynamic response of the TCS module. In this work we aim at understanding the effect of such an additional protein (which we call "third component") on the functional properties of a prototypical TCS. To do so we build mathematical models of TCS with alternative designs for their interaction with that third component. These mathematical models are analyzed in order to identify the differences in dynamic behavior inherent to each design, with respect to functionally relevant properties such as sensitivity to changes in either the parameter values or the molecular concentrations, temporal responsiveness, possibility of multiple steady states, or stochastic fluctuations in the system. The differences are then correlated to the physiological requirements that impinge on the functioning of the TCS. This analysis sheds light on both, the dynamic behavior of synthetically designed TCS, and the conditions under which natural selection might favor each of the designs. We find that a third component that modulates SK activity increases the parameter space where a bistable response of the TCS module to signals is possible, if SK is monofunctional, but decreases it when the SK is bifunctional. The presence of a third component that modulates RR activity decreases the parameter space where a bistable response of the TCS module to signals is possible.     

Cost component analysis

In optimizations the dimension of the problem may severely, sometimes exponentially increase optimization time. Parametric function approximatiors (FAPPs) have been suggested to overcome this problem. Here, a novel FAPP, cost component analysis (CCA) is described. In CCA, the search space is resampled according to the Boltzmann distribution generated by the energy landscape. That is, CCA converts the optimization problem to density estimation. Structure of the induced density is searched by independent component analysis (ICA). The advantage of CCA is that each independent ICA component can be optimized separately. In turn, (i) CCA intends to partition the original problem into subproblems and (ii) separating (partitioning) the original optimization problem into subproblems may serve interpretation. Most importantly, (iii) CCA may give rise to high gains in optimization time. Numerical simulations illustrate the working of the algorithm.     

Principal component approach in variance component estimation for international sire evaluation

Background

The dairy cattle breeding industry is a highly globalized business, which needs internationally comparable and reliable breeding values of sires. The international Bull Evaluation Service, Interbull, was established in 1983 to respond to this need. Currently, Interbull performs multiple-trait across country evaluations (MACE) for several traits and breeds in dairy cattle and provides international breeding values to its member countries. Estimating parameters for MACE is challenging since the structure of datasets and conventional use of multiple-trait models easily result in over-parameterized genetic covariance matrices. The number of parameters to be estimated can be reduced by taking into account only the leading principal components of the traits considered. For MACE, this is readily implemented in a random regression model.

Methods

This article compares two principal component approaches to estimate variance components for MACE using real datasets. The methods tested were a REML approach that directly estimates the genetic principal components (direct PC) and the so-called bottom-up REML approach (bottom-up PC), in which traits are sequentially added to the analysis and the statistically significant genetic principal components are retained. Furthermore, this article evaluates the utility of the bottom-up PC approach to determine the appropriate rank of the (co)variance matrix.

Results

Our study demonstrates the usefulness of both approaches and shows that they can be applied to large multi-country models considering all concerned countries simultaneously. These strategies can thus replace the current practice of estimating the covariance components required through a series of analyses involving selected subsets of traits. Our results support the importance of using the appropriate rank in the genetic (co)variance matrix. Using too low a rank resulted in biased parameter estimates, whereas too high a rank did not result in bias, but increased standard errors of the estimates and notably the computing time.

Conclusions

In terms of estimation''s accuracy, both principal component approaches performed equally well and permitted the use of more parsimonious models through random regression MACE. The advantage of the bottom-up PC approach is that it does not need any previous knowledge on the rank. However, with a predetermined rank, the direct PC approach needs less computing time than the bottom-up PC.     

Methane monooxygenase hydroxylase and B component interactions

The interaction of the soluble methane monooxygenase regulatory component (MMOB) and the active site-bearing hydroxylase component (MMOH) is investigated using spin and fluorescent probes. MMOB from Methylosinus trichosporium OB3b is devoid of cysteine. Consequently, site-directed mutagenesis was used to incorporate single cysteine residues, allowing specific placement of the probe molecules. Sixteen MMOB Cys mutants were prepared and labeled with the EPR spin probe 4-maleimido-2,2,6,6-tetramethyl-1-piperidinyloxy (MSL). Spectral evaluation of probe mobility and accessibility to the hydrophilic spin-relaxing agent NiEDDA showed that both properties decrease dramatically for a subset of the spin labels as the complex with MMOH forms, thereby defining the likely interaction surface on MMOB. This surface contains MMOB residue T111 thought to play a role in substrate access into the MMOH active site. The surface also contains several hydrophilic residues and is ringed by charged residues. The surface of MMOB opposite the proposed binding surface is highly charged, consistent with solvent exposure. Probes of both of the disordered N- and C-terminal regions remain highly mobile and exposed to solvent in the MMOH complex. Spin-labeling studies show that residue A62 of MMOB is located in a position where it can be used to monitor MMOH-MMOB complex formation without perturbing the process. Accordingly, steady-state kinetic assays show that it can be changed to Cys (A62C) and labeled with the fluorescent probes 6-bromoacetyl-2-dimethylaminonaphthalene (BADAN) or 5-((((2-iodoacetyl)amino)ethyl)amino)naphthalene-1-sulfonic acid (1,5-IAEDANS) without loss of the ability of MMOB to promote turnover. The BADAN fluorescence is partially quenched and red shifted as the complex with MMOH forms, allowing affinity measurements. It is shown that the high affinity of labeled MMOB (K(D) = 13.5 nM at pH 6.6, 25 degrees C) for the oxidized MMOH decreases substantially with increasing pH and increasing ionic strength but is nearly unaffected by addition of nonionic detergents. Similarly, the fluorescence anisotropy of the 1,5-IAEDANS-labeled A62C-MMOH complex is perturbed by salts but not nonionic detergents. This suggests that the MMOB-MMOH complex is stabilized by electrostatic interactions consistent with the characteristics of the proposed binding surface. Reduction of MMOH results in a 2-3 order of magnitude decrease in the affinity of the BADAN-labeled A62C-MMOB-MMOH complex, consistent with previous indications of structural change associated with reduction of the active site dinuclear iron cluster. Utilizing BADAN-labeled MMOB, the association and dissociation rate constants for the MMOB-MMOH binding reaction were determined and found to be consistent with a two-step process, possibly involving rapid association followed by a slower conformational change. The latter may be related to the regulation of substrate access into the active site of MMOH.     

Principal component discriminant analysis

The approach adopted involved two-stages. First the 11205 measurements in the mass spectrometry data were reduced to 14 scores by a principal component analysis of the centered but otherwise untreated and unscaled data matrix. Then a linear classifier was derived by linear discriminant analysis using these 14 scores as inputs. This number of scores was chosen by leave-one-out cross-validation on the training set, where it gave an overall error rate of 14%. Some indication of the information used in the classification may be obtained from an inspection of the coefficients of the linear classifier.     

菜豆田叶螨及天敌复合生态系统群落组成及其主成分分析

【目的】叶螨对贵州省菜豆Phaseolus vulgaris的危害日趋严重,为明确菜豆田叶螨及其天敌群落结构特征,对菜豆田叶螨防治提供依据。【方法】于2013年5月至8月对菜豆田叶螨及其天敌进行系统调查,应用群落特征参数、主成分分析法及评价函数对其群落特征、时间格局及群落稳定性进行了研究。【结果】菜豆田叶螨及天敌共有7目11科21种,其中:叶螨4种,优势种为二斑叶螨Tetranychus urticae Koch;捕食性天敌17种,优势种为南方小花蝽Orius similis Zheng。菜豆叶螨及天敌复合系统群落的多样性和均匀度随着植株的生长不断增加,但在植株生长盛期随着二斑叶螨个体数的剧增而下降,在菜豆生育中后期,由于复合生态系统中物种丰富、个体数发生量大,多样性指数和均匀度指数又呈上升趋势。对各时期复合程度的评价表明,中后期复合程度高,初期和末期复合程度低。【结论】复合系统中捕食性天敌对叶螨的控制作用大,优势种天敌表现出一定的跟随效应。     

Multiple group principal component analysis and population differentiation

This paper explores the requirements and advantages of multiple group principal component analysis (MGPCA) when it is used to investigate population differentiation. A distinction is drawn between equality of orientation of the within-group axes and equality of variance along these axes. Several examples of the use of MGPCA are discussed and it is shown that MGPCA per se does not require equality of variance along the axes although it may be a requirement of some of the techniques subsequently used to analyse the component scores. MGPCA is simple and direct, being based on the mathematically well defined eigenvector analysis of a symmetric positive definite (pooled within-group covariance) matrix and it can be thought of as a step in the computation of canonical variate analysis (CVA). It can be used with CVA (which is the most popular method of biometrically assessing population affinities) to assess the contribution of within-group components to among-group discrimination. It is also one of a range of appropriate techniques that can be used to define (and delete if required) within-group growth effects and is particularly suitable when CVA is being used to assess the population affinities. When used in this way it has the advantage of being more influenced by the groups with the greatest growth range.     

Nematicidal activity and chemical component of Poria cocos

Poria cocos, a famous traditional Chinese medicine, was found to have nematicidal activity in experiments searching for nematicidal fungi. The experiment showed it could kill 94.9% of the saprophytic nematode, Panagrellus redivivue, 92.6% of the root-knot nematode, Meloidogyne arenaria, and 93.5% of the pine nematode, Bursaphelenchus xylophilus, on PDA plate within 12 hours. According to the nematicidal activity, three new compounds, 2, 4, 6-triacetylenic octane diacid, 2, 4, 5, 6-tetrahydroxyhexanoic acid and 3, 4-dihydroxy-2-keto-n-butyl 2,4,5,6-tetrahydroxyhexanate, were isolated from submerged cultures of Poria cocos. Of these, 2, 4, 6-triacetylenic octane diacid could kill 83.9% Meloidogyne arenaria and 73.4% Panagrellus redivivus at 500 ppm within 12 hours. Here, it is reported for the first time that Poria cocos has nematicidal activity.