The application of MANOVA to analyse <Emphasis Type="Italic">Arabidopsis thaliana</Emphasis> metabolomic data from factorially designed experiments

Metabolomic technologies produce complex multivariate datasets and researchers are faced with the daunting task of extracting information from these data. Principal component analysis (PCA) has been widely applied in the field of metabolomics to reduce data dimensionality and for visualising trends within the complex data. Although PCA is very useful, it cannot handle multi-factorial experimental designs and, often, clear trends of biological interest are not observed when plotting various PC combinations. Even if patterns are observed, PCA provides no measure of their significance. Multivariate analysis of variance (MANOVA) applied to these PCs enables the statistical evaluation of main treatments and, more importantly, their interactions within the experimental design. The power and scope of MANOVA is demonstrated through two different factorially designed metabolomic investigations using Arabidopsis ethylene signalling mutants and their wild-type. One investigation has multiple experimental factors including challenge with the economically important pathogen Botrytis cinerea and also replicate experiments, while the second has different sample preparation methods and one level of replication ‘nested’ within the design. In both investigations there are specific factors of biological interest and there are also factors incorporated within the experimental design, which affect the data. The versatility of MANOVA is displayed by using data from two different metabolomic techniques; profiling using direct injection mass spectroscopy (DIMS) and fingerprinting using fourier transform infra-red (FT-IR) spectroscopy. MANOVA found significant main effects and interactions in both experiments, allowing a more complete and comprehensive interpretation of the variation within each investigation, than with PCA alone. Canonical variate analysis (CVA) was applied to investigate these effects and their biological significance. In conclusion, the application of MANOVA followed by CVA provided extra information than PCA alone and proved to be a valuable statistical addition in the overwhelming task of analysing metabolomic data.     

Computational tools for isotopically instationary 13C labeling experiments under metabolic steady state conditions

(13)C metabolic flux analysis (MFA) has become an important and powerful tool for the quantitative analysis of metabolic networks in the framework of metabolic engineering. Isotopically instationary (13)C MFA under metabolic stationary conditions is a promising refinement of classical stationary MFA. It accounts for the experimental requirements of non-steady-state cultures as well as for the shortening of the experimental duration. This contribution extends all computational methods developed for classical stationary (13)C MFA to the instationary situation by using high-performance computing methods. The developed tools allow for the simulation of instationary carbon labeling experiments (CLEs), sensitivity calculation with respect to unknown parameters, fitting of the model to the measured data, statistical identifiability analysis and an optimal experimental design facility. To explore the potential of the new approach all these tools are applied to the central metabolism of Escherichia coli. The achieved results are compared to the outcome of the stationary counterpart, especially focusing on statistical properties. This demonstrates the specific strengths of the instationary method. A new ranking method is proposed making both an a priori and an a posteriori design of the sampling times available. It will be shown that although still not all fluxes are identifiable, the quality of flux estimates can be strongly improved in the instationary case. Moreover, statements about the size of some immeasurable pool sizes can be made.     

Quantitative analysis of the transgene variability among primary tobacco transformants

The aim of this study was to develop a model for the quantitative estimation of the genetic and environmental variance components in the first generation (T₁) of transgenic plants, in which the transgene effect is considered as a source of genetic variation. The experimental population consisted of T₁ independent transgenic plants (ITPs). Forty-two ITPs of tobacco were generated, containing a chimaeric gene comprising the cauliflower mosaic virus (CaMV) 35S promoter and the reporter gene -glucuronidase (GUS). From each ITP, four cuttings were grown in a randomized block design, and GUS activity in the leaves was determined. The mean GUS activity of the ITPs ranged from 0.55 to 167.9 pmol MU per mg protein per min. Testing of the statistical assumptions of the model revealed a significant scale effect, resulting from correlation between the intra-ITP variance and the average GUS activity of the ITPs. Log GUS activity (LGA) and power of –0.15 of GUS activity (TGA) scale transformations eliminated the scale effect. For GUS activity, the inter-ITP variance was only 28% of the total variance in the experiment, whereas for LGA and TGA it was 72% and 76%, respectively. The opposite was true for the intra-ITP variance, which was reduced from 58% to 18% and 16%, respectively. The experimental design allowed partitioning of the phenotypic variance in T₁ transgenic plants into genetic and environmental components. According to the original scale GUS activity, most of the phenotypic variance was due to environmental variance; the common tendency to interpret this variance as an outcome of position effect and other genetic changes due to transformation leads to incorrect findings. In the present example, after scale transformation the genetic component was 80% of the phenotypic variance.     

Evidence for growth ofSporothrix schenckii on dead but not on living sphagnum moss

When clinical isolates ofSporothrix schenckii were inoculated onto the apices of living or dead sphagnum moss plants maintained under growth chamber conditions, populations of the fungus, assessed by standard dilution plate methods, increased swiftly up to about 70-fold on moist, dead plants but did not increase on the live moss. Light and scanning electron microscopy revealed fungal growth and sporulation on and within dead plants, but no evidence of either on live plants. These data provide indirect support for the contention thatS. schenckii does not grow on living sphagnum in bogs, but rather that sporotrichosis epidemics associated with sphagnum moss are likely to result from contamination of the dead plants at some point(s) in the chain of events during or after harvest. One practical implication of our results is that precautions should be taken to insure that sphagnum moss is stored dry and that it is not wetted any sooner than necessary before use. We also report here improvement of the Mycose isolation medium by an increase in cycloheximide from 400 to 800 mg/l, chloramphenicol from 50 to 250 mg/l, and the addition of rifampicin at 20 mg/l.     

Comparing G: multivariate analysis of genetic variation in multiple populations

The additive genetic variance–covariance matrix (G) summarizes the multivariate genetic relationships among a set of traits. The geometry of G describes the distribution of multivariate genetic variance, and generates genetic constraints that bias the direction of evolution. Determining if and how the multivariate genetic variance evolves has been limited by a number of analytical challenges in comparing G-matrices. Current methods for the comparison of G typically share several drawbacks: metrics that lack a direct relationship to evolutionary theory, the inability to be applied in conjunction with complex experimental designs, difficulties with determining statistical confidence in inferred differences and an inherently pair-wise focus. Here, we present a cohesive and general analytical framework for the comparative analysis of G that addresses these issues, and that incorporates and extends current methods with a strong geometrical basis. We describe the application of random skewers, common subspace analysis, the 4th-order genetic covariance tensor and the decomposition of the multivariate breeders equation, all within a Bayesian framework. We illustrate these methods using data from an artificial selection experiment on eight traits in Drosophila serrata, where a multi-generational pedigree was available to estimate G in each of six populations. One method, the tensor, elegantly captures all of the variation in genetic variance among populations, and allows the identification of the trait combinations that differ most in genetic variance. The tensor approach is likely to be the most generally applicable method to the comparison of G-matrices from any sampling or experimental design.     

Can whole-drawer images measure up? A reply to Johnson et al. (2013)?

Johnson et al. (2013) found that morphometric measurements of dragonfly wings taken from actual specimens and measurements taken from whole-drawer images of those specimens were equally accurate. We do not believe that their conclusions are justified by their data and analysis. Our reasons are, first, that their study was constrained in ways that restrict the generalisability of their results, but second, and of far greater significance, their statistical approach was entirely unsuited to their data and their results misled them to erroneous conclusions. We offer an alternative analysis of their data as published. Our reanalysis demonstrates, contra Johnson et al., that measurements from scanned images are not a reliable substitute for direct measurement.     

The α-glycerophosphate dehydrogenase (α-gpdh) polymorphism in Drosophila melanogaster: adult survival under temperature stress

Summary As a test of the hypothesis that adult temperature stress is an important component of natural selection maintaining the -gpdh polymorphism, we have looked for differential survival among genotypes subjected to (i) heat shock and (ii) cold shock. Factorial ANOVAR, taking account of genotype, sex and temperature-stress indicated that genotype did not contribute to the variance of survival proportion per vial. We have not therefore found evidence to support our hypothesis. Incidental to the above was a significant sex-temperature interaction. Thus, adult females showed higher survival than males under heat stress, while under cold stress, there was no indication of a survival difference between the sexes.     

Statistical issues in quality control of proteomic analyses: good experimental design and planning

Quality control is becoming increasingly important in proteomic investigations as experiments become more multivariate and quantitative. Quality control applies to all stages of an investigation and statistics can play a key role. In this review, the role of statistical ideas in the design and planning of an investigation is described. This involves the design of unbiased experiments using key concepts from statistical experimental design, the understanding of the biological and analytical variation in a system using variance components analysis and the determination of a required sample size to perform a statistically powerful investigation. These concepts are described through simple examples and an example data set from a 2-D DIGE pilot experiment. Each of these concepts can prove useful in producing better and more reproducible data.     

Phylogeny and classification of the Catantopidae at the tribal level (Orthoptera, Acridoidea)

The grasshopper family Catantopidae is a well-known group, whose members include some of the most notorious agricultural pests. The existing classifications of the family are mostly utilitarian rather than being based on phylogenetic analysis and therefore unable to provide the stability desired for such an economically important group. In the present study, we present the first comprehensive phylogenetic analysis of the family based on morphology. By extensively sampling from the Chinese fauna, we included in the present analysis multiple representatives of each of the previously recognized tribes in the family. In total, we examined 94 genera represented by 240 species and evaluated 116 characters, including 84 for external morphology and 32 for male genitalia. The final matrix consists of 86 ingroup taxa and 88 characters. Our phylogenetic analyses resulted in a high resolution of the basal relationships of the family while showed considerable uncertainty about the relationships among some crown taxa. We further evaluated the usefulness of morphological characters in phylogeny reconstruction of the catantopids by examining character fit to the shortest trees found, and contrary to previous suggestions, our results suggest that genitalia characters are not as informative as external morphology in inferring higher-level relationship. We further suggest that earlier classification systems of grasshoppers in general and Catantopidae in particular most probably consist of many groups that are not natural due the heavy reliance on genitalia features and need to be revised in the light of future phylogenetic studies. Finally, we outlined a tentative classification scheme based on the results of our phylogenetic analysis.     

Semiparametric transformation models for multiple continuous biomarkers in ROC analysis

Recent technological advances continue to provide noninvasive and more accurate biomarkers for evaluating disease status. One standard tool for assessing the accuracy of diagnostic tests is the receiver operating characteristic (ROC) curve. Few statistical methods exist to accommodate multiple continuous‐scale biomarkers in the framework of ROC analysis. In this paper, we propose a method to integrate continuous‐scale biomarkers to optimize classification accuracy. Specifically, we develop semiparametric transformation models for multiple biomarkers. We assume that unknown and marker‐specific transformations of biomarkers follow a multivariate normal distribution. Our models accommodate biomarkers subject to limits of detection and account for the dependence among biomarkers by including a subject‐specific random effect. We also propose a diagnostic measure using an optimal linear combination of the transformed biomarkers. Our diagnostic rule does not depend on any monotone transformation of biomarkers and is not sensitive to extreme biomarker values. Nonparametric maximum likelihood estimation (NPMLE) is used for inference. We show that the parameter estimators are asymptotically normal and efficient. We illustrate our semiparametric approach using data from the Endometriosis, Natural History, Diagnosis, and Outcomes (ENDO) study.     

In vitro propagation and chromosome doubling of a Triticum crassum x Hordeum vulgare intergeneric hybrid

Summary In vitro culture of inflorescence tissue of a Triticum crassum (6x) x Hordeum vulgare cv. Bomi (2x) intergeneric hybrid resulted in the proliferation of totipotent callus from which plants were regenerated. Regeneration was also achieved from immature inflorescence callus of T. crassum but not from H. vulgare. T. crassum x H. vulgare regenerates had a somatic chromosome number of 28, identical to that of the original hybrid. Four chimeric plants with a partially doubled chromosome number were obtained by in vitro colchicine treatment of hybrid callus prior to induction of plant regeneration. All T. crassum regenerates had 35 chromosomes rather than the expected number of 42. Meiotic analysis of a 35-chromosome plant revealed an extremely abnormal meiosis which might be attributable to a complete disturbance in meiotic control system(s) including that of meiotic pairing.Dedicated to Professor Georg Melchers on the occasion of his 75th birthday.Contribution No. 624 Ottawa Research Station, Research Branch, Agriculture Canada, Ottawa, Ont. K1A OC6 (Canada)     

Identification of AFLP molecular markers for resistance against Melampsora larici-populina in Populus

We have identified AFLP markers tightly linked to the locus conferring resistance to the leaf rust Melampsora larici-populina in Populus. The study was carried out using a hybrid progeny derived from an inter-specific, controlled cross between a resistant Populus deltoides female and a susceptible P. nigra male. The segregation ratio of resistant to susceptible plants suggested that a single, dominant locus defined this resistance. This locus, which we have designated Melampsora resistance (Mer), confers resistance against E1, E2, and E3, three different races of Melampsora larici-populina. In order to identify molecular markers linked to the Mer locus we decided to combine two different techniques: (1) the high-density marker technology, AFLP, which allows the analysis of thousands of markers in a relatively short time, and (2) the Bulked Segregant Analysis (BSA), a method which facilitates the identification of markers that are tightly linked to the locus of interest. We analyzed approximately 11,500 selectively amplified DNA fragments using 144 primer combinations and identified three markers tightly linked to the Mer locus. The markers can be useful in current breeding programs and are the basis for future cloning of the resistance gene.     

The “Window <Emphasis Type="Italic">t</Emphasis> test”: a simple and powerful approach to detect differentially expressed genes in microarray datasets

This work focuses on differential expression analysis of microarray datasets. One way to improve such statistical analyses is to integrate biological information in the design of these analyses. In this paper, we will use the relationship between the level of gene expression and variability. Using this biological information, we propose to integrate the information from multiple genes to get a better estimate of individual gene variance, when a small number of replicates are available, to increase the power of the statistical analysis. We describe a strategy named the “Window t test” that uses multiple genes which share a similar expression level to compute the variance which is then incorporated a classic t test. The performances of this new method are evaluated by comparison with classic and widely-used methods for differential expression analysis (the classic Student t test, the Regularized t test (reg t test), SAM, Limma, LPE and Shrinkage t). In each case tested, the results obtained were at least equivalent to the best performing method and, in most cases, outperformed it. Moreover, the Window t test relies on a very simple procedure requiring small computing power compared with other methods designed for microarray differential expression analysis. Electronic Supplementary Material  Supplementary material is available for this article at and is accessible for authorized users.     

Telemetry System for Assessing Jaw-Muscle Function in Free-ranging Primates

In vivo laboratory-based studies describing jaw-muscle activity and mandibular bone strain during mastication provide the empirical basis for most evolutionary hypotheses linking primate masticatory apparatus form to diet. However, the laboratory data pose a potential problem for testing predictions of these hypotheses because estimates of masticatory function and performance recorded in the laboratory may lack the appropriate ecological context for understanding adaptation and evolution. For example, in laboratory studies researchers elicit rhythmic chewing using foods that may differ significantly from the diets of wild primates. Because the textural and mechanical properties of foods influence jaw-muscle activity and the resulting strains, chewing behaviors studied in the laboratory may not adequately reflect chewing behaviors of primates feeding in their natural habitats. To circumvent this limitation of laboratory-based studies of primate mastication, we developed a system for recording jaw-muscle electromyograms (EMGs) from free-ranging primates so that researchers can conduct studies of primate jaw-muscle function in vivo in the field. We used the system to record jaw-muscle EMGs from mantled howlers (Alouatta palliata) at Hacienda La Pacifica, Costa Rica. These are the first EMGs recorded from a noncaptive primate feeding in its natural habitat. Further refinements of the system will allow long-term EMG data collection so that researchers can correlate jaw-muscle function with food mechanical properties and behavioral observations. In addition to furthering understanding of primate feeding biology, our work will foster improved adaptive hypotheses explaining the evolution of primate jaw form.     

Food mixing strategies in the desert locust: effects of phase, distance between foods, and food nutrient content

Food mixing strategies were compared in the cryptically coloured, relatively sedentary `solitarious' and the highly mobile, conspicuously coloured `gregarious' phases of the desert locust, Schistocerca gregaria. Based on phase related differences in behaviour and nutritional regulatory responses, we predicted that solitarious nymphs, compared to gregarious nymphs, would move less between nutritionally complementary foods, particularly as the distance between the foods increased. We manipulated the nutritional composition [protein (p) and digestible carbohydrate (c) content] of two foods in an experimental arena and varied the distance between the foods using a factorial experimental design. Results indicated that in general, solitarious nymphs showed greater fidelity to individual food dishes than did gregarious insects (i.e., they concentrated their feeding mainly on one dish). However, results also demonstrated that for both phases fidelity to a particular food dish increased as the distance between the dishes increased, and that the number of switches between dishes decreased with increasing distance. In the smallest arenas, though, gregarious nymphs switched more frequently between the two food dishes than solitarious nymphs, even when the two dishes contained the same, near-optimal food (p18:c24). When challenged by having the two dishes either placed furthest apart (2 m) or more divergent in nutritional composition (p29:c13 vs. p7:c35), insects of both phases regulated protein intake more strongly than carbohydrate intake, by eating more from the dish containing higher-protein food.     

Random homologous pairing and incomplete sister chromatid alignment are common in angiosperm interphase nuclei

The chromosome arrangement in interphase nuclei is of growing interest, e.g., the spatial vicinity of homologous sequences is decisive for efficient repair of DNA damage by homologous recombination, and close alignment of sister chromatids is considered as a prerequisite for their bipolar orientation and subsequent segregation during nuclear division. To study the degree of homologous pairing and of sister chromatid alignment in plants, we applied fluorescent in situ hybridisation with specific bacterial artificial chromosome inserts to interphase nuclei. Previously we found in Arabidopsis thaliana and in A. lyrata positional homologous pairing at random, and, except for centromere regions, sister chromatids were frequently not aligned. To test whether these features are typical for higher plants or depend on genome size, chromosome organisation and/or phylogenetic affiliation, we investigated distinct individual loci in other species. The positional pairing of these loci was mainly random. The highest frequency of sister alignment (in >93% of homologues) was found for centromeres, some rDNA and a few other high copy loci. Apparently, somatic homologous pairing is not a typical feature of angiosperms, and sister chromatid aligment is not obligatory along chromosome arms. Thus, the high frequency of chromatid exchanges at homologous positions after mutagen treatment needs another explanation than regular somatic pairing of homologues (possibly an active search of damaged sites for homology). For sister chromatid exchanges a continuous sister chromatid alignment is not required. For correct segregation, permanent alignment of sister centromeres is sufficient.     

Behavioral responses of a leaf beetle to injury-related changes in its salicaceous host

Summary Adults of the leaf beetle Plagiodera versicolora Laich, avoided previously injured shoots of Salix alba Tristis in favor of nearby uninjured ones. The response was rapid and persisted for five days of study. During brief periods of observation, the vast majority of beetles engaged in behaviors other than locomotion such as feeding and resting. Furthermore, the proportions of beetles walking were similar on injured and uninjured shoots. It appears that under these experimental conditions previous injury did not greatly increase searching behavior of the beetles. Leaves on previously injured shoots received less subsequent herbivory than uninjured ones. Findings of these experiments are consistent with patterns of herbivory observed in the field and known reproductive responses of beetles to diets of injured leaves. These results provide new information toward understanding relatively low levels of herbivory on individual willow leaves associated with the feeding of P. versicolora.