Application of Multiple Test Procedures to the Combination of Multivariate and Univariate Tests with Varying Variable Sets

The positive ascertainment of location differences in a multivariate comparison of two or more groups gives rise to the question for the contribution of the single variables or of subsets of variables to the multivariate difference. In this paper two methods are proposed to accomplish the original multivariate test by tests in variable subsets or in single variables using a closed test procedure and Holm's procedure, respectively. Both control the multiple level of the whole procedure.     

Abstracts of papers to be presented at the Fiftieth Annual Meeting of the American Association of physical anthropologists Detroit,Michigan April 22–25, 1981

The critical problem confronting all allometric studies is the choice of an appropriate size variable, especially when body mass or some other measure of total size is unavailable. A method proposed by Jolicoeur (1963a,b) claims to generate an internal size variable by a principal components analysis of the covariance matrix of logarithmically transformed data, from which allometric coefficients can be computed. Despite the current popularity of this method, the precise relationship and degree of compatibility between such multivariate coefficients and the exponent of the bivariate power function (Y = βX^α) is unknown. This study evaluates the comparability and interpretability of allometric values computed by Jolicoeur's procedure and by standard bivariate regressions (least squares and major axis). Two primate data sets with known measures of size were utilized for these purposes: (1) longitudinal growth data from radiographs of two species of capuchin monkeys, Cebus apella and Cebus albifrons; and (2) interspecific osteometric data from a series of adult lemurs, species of prosimians from Madagascar. Consistent differences exist between multivariate and bivariate allometric coefficients for both ontogenetic and static data sets. Multivariate analysis underestimated the coefficients in the former and overestimated them in the latter. The internal size variable generated by principal components analysis is clearly not equivalent to, and hence not a suitable substitute for, known measures of size. Moreover, multivariate coefficients are very sensitive to the composition of variables in a given data set; α values of a variable changed appreciably depending on the other variables included in the analysis. The multivariate coefficients are also sample-specific, and provide misleading biological information when compared between samples (e.g., between species of capuchin monkey). For allometric investigations designed to evaluate scaling parameters relative to total size, alternative analytical solutions to the Jolicoeur method should be considered.     

MANOVA multiple comparisons using the generalized step-down procedure

If the variables in MANOVA problem can be arranged according to the order of their importance, then J. ROY'S (1958) step-down procedure may be more appropriate than the conventional invariant inference techniques. However, it may often be possible only to identify subsets such that variables within subsets are equally important and subsets are of unequal importance. In experimental situations, it is common to have a set of variables of primary interest and another of “addon” variables. The step-down reasoning is extended to such cases and a set of simultaneous confidence bounds based upon the procedure which uses the largest root criterion at each stage are derived. The confidence bounds are on all linear functions of means only that do not involve nuisance parameters, and are therefore suitable for studying the configuration of means. This method yields shorter intervals for contrasts among the means of the variables of primary interest compared with the conventional intervals based upon the largest root. The method is illustrated using BARNARD'S data (1935) on skull characters.     

Linking the Physicochemical Properties with the Abundance and Diversity of Rhizospheric Bacterial Population Inhabiting Paddy Soil Based on a Concerted Multivariate Analysis of PCR-DGGE and RISA

To unravel the existence of dominant bacterial population in the paddy fields of Eastern Uttar Pradesh, India and their relation to the prevailing soil physicochemistry using multivariate statistical analyses, a cumulative culture-independent 16S rRNA based Polymerase chain reaction-Denaturing gradient gel electrophoresis (PCR-DGGE) and a 16S-23S ribosomal intergenic spacer analysis (RISA) have been performed. Detrended correspondence analysis (DCA) and principal component analysis (PCA) biplot analyses were used to assess the relation between soil bacterial population and its physicochemistry. DCA analysis exhibited a strong dependence of bacterial existence on the soil physicochemical variables, such as organic matter, total nitrogen, inorganic nutrients, temperatures, and moisture status. Soil dehydrogenase activity (DHA) was assessed to check the metabolic activity of all soil samples which showed a range of 0.012–0.050 nmol TPF g^?1 min^?1 with significant variation (p < 0.01). Out of 96 bands excised, 45 different phylotypes were obtained using both techniques which elucidated the abundance of Cyanobacteria over other soil bacterial population. Scytonema sp., Leptolyngbya sp. and different uncultured cyanobacterial species were the major genera found. Profiling data obtained through PCR-DGGE and RISA were used in alpha diversity and rarefaction curve analysis suggested site 6 (Chandauli) as the most diversity rich site. Thus extensive dataset of weighted and unweighted variables generated through DGGE and RISA coupled with metabolic functioning of soil and multivariate analyses provided an excellent opportunity to map the soil microbial structure in paddy fields and their regulation with existing soil environment.     

Ein fortran-unterprogramm zur qnadratischen optimierung von mixturen

A FORTRAN-Subroutine is given to solve an optimization problem connected with the quadratic canonical mixture polynomial of SCHEFFÉ (1958). Two numerical examples demonstrate the use of the procedure. Some remarks about practical experience with the program in fermentation to optimize the compound of nutrient media are added.     

Spatial ordination of vegetation data using a generalization of Wartenberg's multivariate spatial correlation

Question: Are there spatial structures in the composition of plant communities? Methods: Identification and measurement of spatial structures is a topic of great interest in plant ecology. Univariate measurements of spatial autocorrelation such as Moran's I and Geary's c are widely used, but extensions to the multivariate case (i.e. multi‐species) are rare. Here, we propose a multivariate spatial analysis based on Moran's I (MULTISPATI) by introducing a row‐sum standardized spatial weight matrix in the statistical triplet notation. This analysis, which is a generalization of Wartenberg's approach to multivariate spatial correlation, would imply a compromise between the relations among many variables (multivariate analysis) and their spatial structure (autocorrelation). MULTISPATI approach is very flexible and can handle various kinds of data (quantitative and/or qualitative data, contingency tables). A study is presented to illustrate the method using a spatial version of Correspondence Analysis. Location: Territoire d'Etude et d'Expérimentation de Trois‐Fontaines (eastern France). Results: Ordination of vegetation plots by this spatial analysis is quite robust with reference to rare species and highlights spatial patterns related to soil properties.     

Salinivibrio costicola subsp. alcaliphilus subsp. nov., a haloalkaliphilic aerobe from Campania Region (Italy)

Phenotypic and phylogenetic studies were performed on unidentified Gram-negative staining, haloalkaliphilic aerobe and protease producer Salinivibrio-like organism recovered from a saltish spring with algal mat in the “Pozzo del Sale” site (Salt's Well) in the Campania Region (South Italy). Phylogenetic analysis based on comparison of 16S rRNA gene sequences demonstrated that the isolate was related to species of Salinivibrio genus. The DNA–DNA hybridization of the type strain 18AG^T with the most related Salinivibrio costicola subsp. costicola showed a re-association value of 72%. Based on the phenotypic distinctiveness of 18AG^T strain and molecular, chemical and genetic evidence, it is proposed that strain 18AG^T can be classified as S. costicola subsp. alcaliphilus, subsp. nov. The type strain of S. costicola subsp. alcaliphilus, is ATCC BAA-952^T; DSM 16359^T.     

Directional Tests for the Analysis of Clinical Trials with Multiple Endpoints Allowing for Incomplete Data

Incomplete data are a serious problem in the multivariate analysis of clinical trials. Usually a complete-case analysis is performed: All incomplete observation vectors are excluded from the analysis. Provided that observations are missing randomly, an easy-to-handle available-case analysis is introduced, allowing the analysis of all data without insertion or deletion of observations. This method is applied to parametric and nonparametric test procedures of the O'Brien type, which are more powerful than the conventional Hotelling's T² for detecting alternatives where the (treatment) effect has the same direction for all observed variables. In addition, the applicability of these so-called directional tests, especially in the case of small samples, and their pros and cons are discussed.     

基于高光谱数据的互花米草叶片功能性状反演

互花米草成功入侵的关键是其生长繁殖能力以及对环境的适应能力,叶片含水率、相对叶绿素含量、碳氮比、总氮、总磷以及比叶面积等叶片功能性状反应的是互花米草对资源的利用能力以及环境的适应能力。以江苏盐城滨海湿地为研究对象,进行互花米草叶片功能性状与高光谱数据的关系研究。通过对原始光谱数据以及一阶微分转换光谱数据进行主成分分析提取新的主成分变量作为自变量分别建立不同性状的逐步回归、BP神经网络、支持向量机、随机森林4种预测模型,通过比较构建模型的R²以及RMSE选择最优模型,进而基于相关性分析得到的敏感波段构建最优模型,验证其准确性和适用性。研究结果发现：（1）一阶微分数据的建模效果优于原始光谱数据;（2）通过对不同功能性状的预测建模,发现4种模型的预测效果排序为：随机森林>支持向量机>BP神经网络>逐步回归,其中随机森林模型的准确性高、稳定性强,明显优于其他3种模型,而逐步回归模型的效果最差,不适用于互花米草叶片功能性状的高光谱建模;（3）通过对相关性分析得到的敏感波段建立随机森林模型,建模R²均大于0.90,验证R²介于0.73-0.95之间,进一步证实了随机森林模型的准确性和稳定性。研究结果表明,高光谱数据可以作为快速监测互花米草生长状况的有力手段,而随机森林模型可以作为高精度模型实现对互花米草不同叶片功能性状的估测。     

A remark on a paper by misra

MISRA (1978) sets confidence intervals for a double linear compound of multivariate normal regression coefficients by using ROY'S maximum root test criterion. The exact test statistic to be used is STUDENT'S t. The t statistic gives narrower confidence bounds than those given by ROY's maximum root statistic. A result given by MORRISON (1975, p. 18, equation 10) for profile analysis is also obtained by using the STUDENT'S t test.     

Concurrent Management of an Exotic Species and Initial Restoration Efforts in Forests

One of the proximate results of forest fragmentation, and a cause of continued microenvironmental change and exacerbation of ecological problems, is increased invasions by weedy plant species. One such example is Alliaria petiolata (Brassicaceae), a serious pest threatening much of eastern North America. Alliaria petiolata impedes mitigation of fragmentation and restoration efforts because it tends to outcompete and possibly extirpate much of the native understory species on localized scales. As part of a strategy to address the problems of fragmented habitats, an experiment was conducted to determine whether Sanguinaria canadensis (Papaveraceae) could outcompete A. petiolata. Using an additive design, I transplanted S. canadensis at densities of 0, 1, 2, 3, 5, 7, 9, 11, 15, and 20 ramets/m² in 1997 and allowed them to interact with initial A. petiolata densities of 128 seedlings and 31 rosettes/m². As of 2000, multivariate analyses of variance with repeated measures and simple analyses of variance indicated that initial S. canadensis densities of as little as 5 ramets/m² suppressed A. petiolata. Initial S. canadensis densities of 9 and 11 ramets/m² resulted in the lowest numbers of late‐spring seedlings, numbers and sizes of year 1 and 2 rosettes, numbers and gross areas of stem leaves, numbers of flowering individuals, number of flowers, number of fruits (siliques), and height at flowering. While it remains to be tested whether this will continue and if the reestablishment of S. canadensis will help reassemble forest ecosystems, the experiments indicated that transplanting S. canadensis was effective at mitigating the spread of A. petiolata.     

高分辨魔角旋转核磁共振和主成分分析研究人类低级星形细胞瘤和脑膜瘤的代谢组特征

采用高分辨魔角旋转核磁共振（HRMAS ^1H NMR）技术结合主成分分析（PCA）方法研究了39例人体脑肿瘤组织的代谢组特征．39例肿瘤样本分别来自39个脑肿瘤患者,包括15例低级星形细胞瘤,13例纤维型脑膜瘤和11例过渡型脑膜瘤．核磁共振波谱分析结果表明,脑肿瘤组织的代谢组中丰要含有脂肪酸、乳酸、胆碱代谢物（如胆碱、磷酸胆碱和甘油磷酸胆碱）、氯基酸（如丙氨酸、谷氨酸、谷氮酰胺、牛磺酸）、N-乙酰天门冬氨酸（NAA）和谷胱甘肽等代谢物．通过对核磁共振谱进行主成分分析（PCA）,发现低级星形细胞瘤和脑膜瘤的代谢组之间具有明显的差异,而在过渡型和纤维型两个亚类脑膜瘤之间该差别相对较小．与脑膜瘤相比,低级星形细胞瘤中甘油磷酸胆碱、磷酸胆碱、肌醇与肌酸的含量较高,而丙氨酸、谷氨酸、谷氨酰胺、谷胱甘肽和牛磺酸的含量较低．NAA的含量在低级星形细胞瘤中尽管较低但能观察到,而脑膜瘤中却未发现NAA的信号．结果衷明,HRMAS ^1H NMR和多变量统计分析相结合的组织代谢组学方法,不仅能有效区分不同类型的脑肿瘤,而且还可以为脑肿瘤提供丰富的代谢组信息,这些信息对研究肿瘤发生发展的机制具有潜在的意义．     

Protein engineering of toluene-o-xylene monooxygenase from Pseudomonas stutzeri OX1 for enhanced chlorinated ethene degradation and o-xylene oxidation

Toluene-o-xylene monooxygenase (ToMO) from Pseudomonas stutzeri OX1 has been shown to degrade all chlorinated ethenes individually and as mixtures. Here, DNA shuffling of the alpha hydroxylase fragment of ToMO (TouA) and saturation mutagenesis of the TouA active site residues I100, Q141, T201, F205, and E214 were used to enhance the degradation of chlorinated aliphatics. The ToMO mutants were identified using a chloride ion screen and then were further examined by gas chromatography. Escherichia coli TG1/pBS(Kan)ToMO expressing TouA saturation mutagenesis variant I100Q was identified that has 2.8-fold better trichloroethylene (TCE) degradation activity (apparent V _max of 1.77 nmol min⁻¹ mg⁻¹ protein⁻¹ vs 0.63 nmol min⁻¹ mg⁻¹ protein⁻¹). Another variant, E214G/D312N/M399V, has 2.5-fold better cis-1,2-dichloroethylene (cis-DCE) degradation activity (apparent V _max of 8.4 nmol min⁻¹ mg⁻¹ protein⁻¹ vs 3.3 nmol min⁻¹ mg⁻¹ protein⁻¹). Additionally, the hydroxylation regiospecificity of o-xylene and naphthalene were altered significantly for ToMO variants A107T/E214A, T201G, and T201S. Variant T201S produced 2.0-fold more 2,3-dimethylphenol (2,3-DMP) from o-xylene than the wild-type ToMO, whereas variant A107T/E214A had 6.0-fold altered regiospecificity for 2,3-DMP formation. Variant A107T/E214A also produced 3.0-fold more 2-naphthol from naphthalene than the wild-type ToMO, whereas the regiospecificity of variant T201S was altered to synthesize 3.0-fold less 2-naphthol, so that it made almost exclusively 1-naphthol (96%). Variant T201G was more regiospecific than variants A107T/E214A and T201S and produced 100% 3,4-DMP from o-xylene and >99% 1-naphthol from naphthalene. Hence, ToMO activity was enhanced for the degradation of TCE and cis-DCE and for the regiospecific hydroxylation of o-xylene and naphthalene through DNA shuffling and saturation mutagenesis.     

Efficiency of a Two-Stage Procedure for Survival Studies

A two-stage procedure using occurrences of a surrogate endpoint regarding a true clinical endpoint has been proposed (FLANDRE, O'QUIGLEY and BROOKMEYER, 1994). The procedure allows estimation of the survival function S(t) reliably and comparison of treatment groups. The aim of this paper is to investigate the efficiency of such a two-stage procedure compared to a one stage procedure, i.e. a classical study. The efficiency is estimated by the ratio of the standard derivation of S(t) for the classical procedure to the standard derivation of S(t) for the two-stage procedure and represents the precision on the estimation of S(t). Standard deviation of S(t) is approximate by the delta-method. Using the exponential model. some numerical results are presented.     

Bending of z-lines by mechanical stimuli: An input signal for integrin dependent modulation of ion channels?

We studied which components of mechanical cell deformation are involved in “stretch modulated ion currents” (SMIC). Murine ventricular myocytes were attached to glass coverslips and deformed in x, y and z with a 16 μm thin glass stylus (S) of calibrated stiffness. Three-dimensional confocal microscopy characterized cell deformation (T-tubular membranes, mitochondria) and bending of S (indicative of the applied force). Axial (x-) displacement of S sheared the upper cell part versus the attached bottom, close to S, it changed sarcomere length and bent z-lines (“z-line displacement”). Vertical (z-press) or transversal (y-shear) displacement of S bulged cytoplasm and mitochondria transversally without detectable z-line displacement.Axial stiffness increased with the extent of stress (“stress stiffening”). Depolymerization of F-actin or block of integrin receptors reduced stiffness. SMIC served as a proxy readout of deformation-induced signaling. Axial deformation activated a non-selective cation conductance (G_ns) and deactivated an inwardly rectifying K⁺ conductance (G_K1), z-press or y-shear did not induce SMIC. Depolymerization of F-actin or block of integrin receptors reduced SMIC. SMIC did not depend on changes in sarcomere length but correlated with the extent of z-line bending. We discuss that both shear stress at the attached cell bottom and z-line bending could activate mechanosensors. Since SMIC was absent during deformations without z-line bending we postulate that z-line bending is a necessary component for SMIC signaling.     

Litter-associated bacteria and fungi – a comparison of biomass and communities across lakes and plant species

1. We investigated the importance of lake water chemistry and substrate properties in regulating microbial decomposer communities on macrophyte litter. Ten lakes of differing water chemistry, including such variables as nutrient concentration, pH and dissolved organic carbon, were sampled in October 2003. Water chemistry was analysed, and litter from the macrophytes Phragmites australis, Schoenoplectus lacustris and Nuphar lutea was collected from both above and below the water surface. 2. The three plant species differed widely in carbon and nitrogen content. The aerial parts of P. australis had highest C : N ratio (mean value 125), while the lowest values were found in leaves of N. lutea (19.8). 3. Fungal carbon ranged from 0.15 to 6.4 mg g^?1 dry weight (DW), and was higher on aerial than on submerged plant parts. Fungal biomass was highest on S. lacustris and lowest on N. lutea. Denaturing gradient gel electrophoresis revealed no differences in the number of fungal taxa between plant species or plant parts, with the limitation that no molecular analysis was possible for N. lutea. 4. In contrast, bacteria were most abundant on N. lutea, but showed no significant difference between leaf and stem parts. The number of bacterial taxa was highest on the submerged parts of P. australis. 5. The correlations between microbial variables and the properties of lakes and litter were analysed using multivariate statistics. In a principal component analysis, litter properties explained 78% of the variation in microbial variables. In contrast, redundancy analysis revealed that the explanatory power of lake water chemistry was only 20%, indicating that the properties of the growth substratum were more important than those of the lake water for the attached microbial communities.     

An artifact in measurement of S phase initiation and its implication for the kinetics of S phase-specific enzyme activities

It is shown that the different onset of S phase as measured by autoradiography vs cumulative thymidine uptake is an artifact. We consequently propose that S phase-specific enzyme activities may accumulate a few hours prior to the actual initiation of DNA synthesis. A “pre-S” DNA synthesis that can be readily detected only by autoradiography has been proposed. Published data show that DNA synthesis in cultured animal cells is initiated approx. 2 h later when measured by cumulative incorporation of [³H]thymidine ([³H]TdR) as compared with autoradiography. We show here that the difference is in reality an artifact, owing to not taking into account both gradual, asynchronous entry of cells into S phase, as well as time-dependent accumulation of radioactivity into each cell after it has entered S phase. Combination of these two factors leads to the conclusion that [³H]TdR should be incorporated approximately as the square of time following entry of the first cell into S. Taking this into account, the two methods then are in agreement, as predicted. This argument also applies to the enzyme activities shown to increase with DNA synthesis in synchronized cultures. Such an enzyme accumulation really could begin some time earlier than indicated by conventional plots of cumulative enzyme activity vs time and may, in fact, precede the onset of S by a few hours.     

A revision of chiggers of the <Emphasis Type="Italic">minuta</Emphasis> species-group (Acari: Trombiculidae: <Emphasis Type="Italic">Neotrombicula</Emphasis> Hirst, 1925) using multivariate morphometrics

We revise chiggers belonging to the minuta-species group (genus Neotrombicula Hirst, 1925) from the Palaearctic using size-free multivariate morphometrics. This approach allowed us to resolve several diagnostic problems. We show that the widely distributed Neotrombicula scrupulosa Kudryashova, 1993 forms three spatially and ecologically isolated groups different from each other in size or shape (morphometric property) only: specimens from the Caucasus are distinct from those from Asia in shape, whereas the Asian specimens from plains and mountains are different from each other in size. We developed a multivariate classification model to separate three closely related species: N. scrupulosa, N. lubrica Kudryashova, 1993 and N. minuta Schluger, 1966. This model is based on five shape variables selected from an initial 17 variables by a best subset analysis using a custom size-correction subroutine. The variable selection procedure slightly improved the predictive power of the model, suggesting that it not only removed redundancy but also reduced ‘noise’ in the dataset. The overall classification accuracy of this model is 96.2, 96.2 and 95.5%, as estimated by internal validation, external validation and jackknife statistics, respectively. Our analyses resulted in one new synonymy: N. dimidiata Stekolnikov, 1995 is considered to be a synonym of N. lubrica. Both N. scrupulosa and N. lubrica are recorded from new localities. A key to species of the minuta-group incorporating results from our multivariate analyses is presented.