Skip to main content

METODOS ESTADISTICOS EN ESPACIOS RESTRINGIDOS

MTM2012-33236

•
Funding agency name Ministerio de Economía y Competitividad
Funding agency acronym MINECO
Program Programa Nacional de Proyectos de Investigación Fundamental
Subprogram Subprograma de Proyectos de Investigación Fundamental no Orientada
Call Proyectos I+D
Call year 2012
State Plan 2008-2011 Plan Nacional de Investigación Científica, Desarrollo e Innovación Tecnológica
Management unit Secretaría de Estado de Investigación, Desarrollo e Innovación
Beneficiary centre UNIVERSITAT DE GIRONA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 26
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Proportionality: a valid alternative to correlation for relative data

  • Lovell, David
  • Pawlowsky-Glahn, Vera
  • Egozcue Rubí, Juan José|||0000-0002-5144-4483
  • Marguerat, Samuel
  • Baehler, Juerg
In the life sciences, many measurement methods yield only the relative abundances of different components in a sample. With such relative-or compositional-data, differential expression needs careful interpretation, and correlation-a statistical workhorse for analyzing pairwise relationships-is an inappropriate measure of association. Using yeast gene expression data we show how correlation can be misleading and present proportionality as a valid alternative for relative data. We show how the strength of proportionality between two variables can be meaningfully and interpretably described by a new statistic. which can be used instead of correlation as the basis of familiar analyses and visualisation methods, including co-expression networks and clustered heatmaps. While the main aim of this study is to present proportionality as a means to analyse relative data, it also raises intriguing questions about the molecular mechanisms underlying the proportional regulation of a range of yeast genes., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Differential effects of genetic vs. environmental quality in Drosophila melanogaster suggest multiple forms of condition dependence

  • Bonduriansky, Russell
  • Mallet, Martin A.
  • Arbuthnott, Devin
  • Pawlowsky-Glahn, Vera
  • Egozcue, Juan José
  • Rundle, Howard D.
Condition is a central concept in evolutionary ecology, but the roles of genetic and environmental quality in condition-dependent trait expression remain poorly understood. Theory suggests that condition integrates genetic, epigenetic and somatic factors, and therefore predicts alignment between the phenotypic effects of genetic and environmental quality. To test this key prediction, we manipulated both genetic (mutational) and environmental (dietary) quality in Drosophila melanogaster and examined responses in morphological and chemical (cuticular hydrocarbon, CHC) traits in both sexes. While the phenotypic effects of diet were consistent among genotypes, effects of mutation load varied in magnitude and direction. Average effects of diet and mutation were aligned for most morphological traits, but non-aligned for the male sexcombs and CHCs in both sexes. Our results suggest the existence of distinct forms of condition dependence, one integrating both genetic and environmental effects and the other purely environmental. We propose a model to account for these observations, We thank M. Adler and L. Rowe for discussion, and A. Moore, W. Forstmeier and an anonymous reviewer for thoughtful comments on the manuscript. MAM created the hemiclonal lines and performed mutation accumulation in the laboratory of, and supported by, A. Chippindale. E. Bath, E. Cassidy and M. Telford assisted with morphometrics. Funding was provided by the Australian Research Council (RB), the Natural Sciences and Engineering Research Council of Canada (DA, HDR), the Canadian Foundation for Innovation (HDR) and the Spanish Ministry of Economy and Competitiveness under the project 'METRICS' (Ref. MTM2012-33236)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Modelling of weather parameters to predict russet on ‘Golden Delicious’ apple

  • Barceló i Vidal, Carles
  • Bonany, J.
  • Martín Fernández, Josep Antoni
  • Carbó, J.
Russet on ‘Golden Delicious’ apple (Malus × domestica Borkh.) fruit is a physiological disorder that causes major economic losses to growers. Large variations occur in the severity of russet from one year to another. In Girona (Spain), good correlations were found between the annual severity of russet at harvest and several weather parameters measured shortly after full bloom. A specific statistical methodology for the analysis of compositional data (CoDa) was used to establish these correlations. The most important factor was the percentage of time at relative humidity values > 55% from 30 – 34 d after full bloom (DAFB), which yielded a high correlation (R = 0.80). The percentage of rainy days from 0 – 34 DAFB was also positively correlated with the severity of russet (R = 0.80). Ordinal logit regression models that included these two climatic variables strongly predicted a low, moderate, or high annual severity of russet. Understanding the effects of weather on russet, and developing predictive models may help to manage the marketing of this apple variety which is prone to russet in some areas of cultivation, This research was supported by the Spanish Ministry of Science and Innovation under Projects MTM2009 13272 and MTM2012-33236, and by the Agència de Gestió d’Ajuts Universitaris i de Recerca of the Generalitat de Catalunya (Ref. 2009SGR424)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Proportionality: a valid alternative to correlation for relative data

  • Lovell, David
  • Pawlowsky-Glahn, Vera
  • Egozcue Rubí, Juan José
  • Marguerat, Samuel
  • Baehler, Juerg
In the life sciences, many measurement methods yield only the relative abundances of different components in a sample. With such relative-or compositional-data, differential expression needs careful interpretation, and correlation-a statistical workhorse for analyzing pairwise relationships-is an inappropriate measure of association. Using yeast gene expression data we show how correlation can be misleading and present proportionality as a valid alternative for relative data. We show how the strength of proportionality between two variables can be meaningfully and interpretably described by a new statistic. which can be used instead of correlation as the basis of familiar analyses and visualisation methods, including co-expression networks and clustered heatmaps. While the main aim of this study is to present proportionality as a means to analyse relative data, it also raises intriguing questions about the molecular mechanisms underlying the proportional regulation of a range of yeast genes., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Vegetation patterns in the Southern Apennines (Italy) during MIS 13: Deciphering pollen variability along a NW-SE transect

  • Ermolli, Elda Russo
  • Di Donato, Valentino
  • Martín Fernández, Josep Antoni
  • Orain, Ronan
  • Lebreton, Vincent
  • Piovesan, Gianluca
Pollen data from five palaeo-lacustrine basins are used in this study to correlate floral and vegetation patterns along a NW-SE transect in the Southern Apennines (Italy) during Marine Isotope Stage (MIS) 13. Compositional Data Analysis (CoDA) is applied to pollen data in order to better define and explain the main differences in the pollen composition of coeval records. Pollen data belonging to MIS 13, chosen on the basis of a constrained cluster analysis on each entire record, have been analyzed through a non-constrained cluster analysis. The relative variation biplot, the parametric and the non-parametric MANOVA computed for the MIS 13 pollen data, show that samples belonging to the different sites are distinct by their average composition. This means that each site has a distinctive character of forest landscape, which could have been the result of local environmental (physical) factors, such as mesoclimate, topography and edaphic humidity, and/or of historical factors such as species composition of refuges and temporary changes in environmental conditions due to disturbance ecology, This research has been supported by the Spanish Ministry of Economy and Competitiveness under the project "METRICS" Ref. MTM2012-33236. The authors wish to thank two referees and the editor for their helpful comments
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Cokriging of compositional balances including a dimension reduction and retrieval of original units

  • Pawlowsky-Glahn, Vera
  • Egozcue, Juan José
  • Olea, Ricardo A.
  • Pardo-Igúzquiza, Eulogio
Compositional data constitutes a special class of quantitative measurements involving parts of a whole. The sample space has an algebraic-geometric structure different from that of real-valued data. A subcomposition is a subset of all possible parts. When compositional data values include geographical locations, they are also regionalized variables. In the Earth sciences, geochemical analyses are a common form of regionalized compositional data. Ordinarily, there are measurements only at data locations. Geostatistics has proven to be the standard for spatial estimation of regionalized variables but, in general, the compositional character of the geochemical data has been ignored. This paper presents in detail an application of cokriging for the modelling of compositional data using a method that is consistent with the compositional character of the data. The uncertainty is evaluated by a Monte Carlo procedure. The method is illustrated for the contents of arsenic and iron in groundwaters in Bangladesh, which have the peculiarity of being measured in milligrams per litre, units for which the sum of all parts does not add to a constant. Practical results include maps of estimates of the geochemical elements in the original concentration units, as well as measures of uncertainty, such as the probability that the concentration may exceed a given threshold. Results indicate that probabilities of exceedance in previous studies of the same data are too low, This research has been partly supported by the Spanish Ministry of Economy and Competitiveness under the project METRICS (Ref. MTM2012-33236); and by the Agencia de Gestió d’Ajuts Universitaris i de Recerca of the Generalitat de Catalunya under project Ref: 2009SGR424
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Bayes hilbert spaces

  • Boogaart, K. Gerald van den
  • Egozcue, Juan José
  • Pawlowsky-Glahn, Vera
A Bayes linear space is a linear space of equivalence classes of proportional σ-finite measures, including probability measures. Measures are identified with their density functions. Addition is given by Bayes' rule and substraction by Radon-Nikodym derivatives. The present contribution shows the subspace of square-log-integrable densities to be a Hilbert space, which can include probability and infinite measures, measures on the whole real line or discrete measures. It extends the ideas from the Hilbert space of densities on a finite support towards Hilbert spaces on general measure spaces. It is also a generalisation of the Euclidean structure of the simplex, the sample space of random compositions. In this framework, basic notions of mathematical statistics get a simple algebraic interpretation. A key tool is the centred-log-ratio transformation, a generalization of that used in compositional data analysis, which maps the Hilbert space of measures into a subspace of square-integrable functions. As a consequence of this structure, distances between densities, orthonormal bases, and Fourier series representing measures become available. As an application, Fourier series of normal distributions and distances between them are derived, and an example related to grain size distributions is presented. The geometry of the sample space of random compositions, known as Aitchison geometry of the simplex, is obtained as a particular case of the Hilbert space when the measures have discrete and finite support, This research was supported by the Spanish Ministries of Education and Science and of Economy and Competitiveness under three projects: 'Ingenio Mathematica (i-MATH)' Ref. No. CSD2006-00032; 'CODA-RSS' Ref. MTM2009-13272; and 'METRICS', Ref. MTM2012-33236. It was also supported by the Agencia de Gestio d'Ajuts Universitaris i de Recerca of the Generalitat de Catalunya under the project Ref: 2009SGR424
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Signal Interpretation in Hotelling's T2 Control Chart for Compositional Data

  • Vives Mestres, Marina
  • Daunis-i-Estadella, Pepus
  • Martín Fernández, Josep Antoni
Nowadays, control of concentrations of elements is of crucial importance in industry. Concentrations are expressed in terms of proportions or percentages which means that they are compositional data (CoDa). CoDa are defined as vectors of positive elements that represent parts of a whole and usually add to a constant sum. Classical T2 control chart is not appropriate for CoDa, for which is better to use a compositional T2 control chart (T2C CC). This paper generalizes the interpretation of the out-of-control signals of the individual T2C CC for more than three components. We propose two methods for identifying the ratio of components that mainly contribute to the signal. The first one is suitable for low dimensional problems and consists on finding the log ratio of components that maximizes the univariate T2 statistic. The second one is an optimized method for large dimensional problems that simplifies the calculus by transforming the coordinates into the sphere. We illustrate the T2C CC signal interpretation with a practical example from the chemical and pharmaceutical industry, This work has been partially financed by the Ministerio de Ciencia e Innovaci´on (Ref: MTM2012- 33236) and the Ag`encia de Gesti´o d’Ajuts Universitaris i de Recerca (AGAUR), Generalitat de Catalunya (Ref: 2014SGR551)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

On the interpretation of differences between groups for compositional data

  • Martín Fernández, Josep Antoni
  • Daunis-i-Estadella, Pepus
  • Mateu i Figueras, Glòria
Social polices are designed using information collected in surveys; such as the Catalan TimeUse survey. Accurate comparisons of time use data among population groups are commonlyanalysed using statistical methods. The total daily time expended on different activities by asingle person is equal to 24 hours. Because this type of data are compositional, its sample spacehas particular properties that statistical methods should respect. The critical points required tointerpret differences between groups are provided and described in terms of log-ratio methods.These techniques facilitate the interpretation of the relative differences detected in multivariateand univariate analysis, This research was supported by the Ministerio de Econom´ıa y Competividad under the project “METRICS” Ref. MTM2012-33236; and by the Ag`encia de Gesti ´o d’Ajuts Universitaris i de Recerca of the Generalitat de Catalunya under the project Ref: 2014SGR551
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

zCompositions - R package for multivariate imputation of left-censored data under a compositional approach

  • Palarea Albaladejo, Javier
  • Martín Fernández, Josep Antoni
zCompositions is an R package for the imputation of left-censored data under a compositional approach. It is pertinent when the analyst assumes that the relevant information is contained on the relative variation structure of the data. For instance, in cases where the experimental data are simultaneously measured in amounts related to a same total weight or volume. The approach is used in fields like geochemistry of waters or sedimentary rocks, environmental studies related to air pollution, physicochemical analysis of glass fragments in forensic science, and among many others. In these fields, rounded zeros and nondetects are usually regarded as left-censored data that hamper any subsequent data analysis. The implemented methods consider aspects of relevance for a compositional approach such as scale invariance, subcompositional coherence or preserving the multivariate relative structure of the data. Based on solid statistical frameworks, it comprises the ability to deal with single and varying censoring thresholds, consistent treatment of closed and non-closed data, exploratory tools, multiple imputation, MCMC, robust and non-parametric alternatives, and recent proposals for count data. Key methodological aspects, new contributions, computational implementation and the practical application of the approach are discussed, This research has been partially supported by the Scottish Government's Rural and Environment Science and Analytical Services Division (RESAS), the Spanish Ministry of Economy and Competitiveness under the project "METRICS" (Ref. MTM2012-33236) and the Agencia de Gestio d'Ajuts Universitaris i de Recerca (AGAUR), Generalitat de Catalunya (Ref: 2014SGR551)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Independence in Contingency Tables Using Simplicial Geometry

  • Egozcue, Juan José
  • Pawlowsky-Glahn, Vera
  • Templ, Matthias
  • Hron, Karel
Frequently, contingency tables are generated in a multinomial sampling. Multinomial probabilities are then organized in a table assigning probabilities to each cell. A probability table can be viewed as an element in the simplex. The Aitchison geometry of the simplex identifies independent probability tables as a linear subspace. An important consequence is that, given a probability table, the nearest independent table is obtained by orthogonal projection onto the independent subspace. The nearest independent table is identified as that obtained by the product of geometric marginals, which do not coincide with the standard marginals, except in the independent case. The original probability table is decomposed into orthogonal tables, the independent and the interaction tables. The underlying model is log-linear, and a procedure to test independence of a contingency table, based on a multinomial simulation, is developed. Its performance is studied on an illustrative example, This research was supported by thegrant COST Action CRoNoS IC1408, the grant IGA PrF 2015 013 Mathematical Models of the Internal Grant Agency of the Palacky University in Olomouc and the Spanish Ministries of Science and Innovation and Economy and Competitivity under the projects CODA-RSS Ref. MTM2009-13272 and METRICS Ref. MTM2012-33236
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

A bootstrap estimation scheme for chemical compositional data with nondetects

  • Palarea Albaladejo, Javier
  • Martín Fernández, Josep Antoni
  • Olea, Ricardo A.
The bootstrap method is commonly used to estimate the distribution of estimators and their associated uncertainty when explicit analytic expressions are not available or are difficult to obtain. It has been widely applied in environmental and geochemical studies, where the data generated often represent parts of whole, typically chemical concentrations. This kind of constrained data is generically called compositional data, and they require specialised statistical methods to properly account for their particular covariance structure. On the other hand, it is not unusual in practice that those data contain labels denoting nondetects, that is, concentrations falling below detection limits. Nondetects impede the implementation of the bootstrap and represent an additional source of uncertainty that must be taken into account. In this work, a bootstrap scheme is devised that handles nondetects by adding an imputation step within the resampling process and conveniently propagates their associated uncertainly. In doing so, it considers the constrained relationships between chemical concentrations originated from their compositional nature. Bootstrap estimates using a range of imputation methods, including new stochastic proposals, are compared across scenarios of increasing difficulty. They are formulated to meet compositional principles following the log-ratio approach, and an adjustment is introduced in the multivariate case to deal with nonclosed samples. Results suggest that nondetect bootstrap based on model-based imputation is generally preferable. A robust approach based on isometric log-ratio transformations appears to be particularly suited in this context. Computer routines in the R statistical programming language are provided, This research has been supported by the Scottish Government's Rural and Environment Science and Analytical Services Division (RESAS), the Spanish Ministry of Economy and Competitiveness under the project 'METRICS' Ref. MTM2012-33236 and the Agencia de Gestio d'Ajuts Universitaris i de Recerca of the Generalitat de Catalunya under the project Ref. 2009SGR424
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Out-of-control signals in three-part compositional T2 control chart

  • Vives Mestres, Marina
  • Daunis-i-Estadella, Pepus
  • Martín Fernández, Josep Antoni
nterpretation of out-of-control signals in multivariate control charts is a persistent problem. Identifying the variables contributing to the signal is of crucial importance for process control. Both procedures turn out to be more complicated when observations are compositions (variables adding to a constant sum). Compositional T-2 control chart (TC2) is suitable for monitoring compositions because it is based on a transformation of the data, which moves it from restricted to real space. This paper proposes a method for interpreting the out-of-control signals of the TC2 control chart for three-part compositions on the basis of an appropriate selection of the transformation. An example of industrial application illustrates the introduced techniques, This work has been partially financed by the Agencia de Gestio d'Ajuts Universitaris i de Recerca (Ref: 2009SGR424) and by the Ministerio de Ciencia e Innovacion (Ref: MTM2012-33236 and Ref: MTM2009-13272). The authors wish to thank the Universitat de Girona for the grant given to Marina Vives-Mestres during her PhD studies (BR2011/66)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Size Fraction Effects on Planktonic Foraminifera Assemblages: A Compositional Contribution to the Golden Sieve Rush

  • Di Donato, Valentino
  • Daunis-i-Estadella, Pepus
  • Martín Fernández, Josep Antoni
  • Esposito, Paola
A comparative study of Last Glacial to Holocene planktonic foraminiferal assemblages obtained from 150 to 106 m size fractions of a Tyrrhenian sea core was carried out by means of compositional data analysis methods. The results show a major loss of small-sized species in the 150 m size fraction, resulting in underdetection limit values for several Holocene samples. Statistical tests, performed by means of contrast and bootstrap, indicate that the assemblages obtained differ statistically. Nevertheless, despite the drawbacks of the analysis based on the 150 m fraction, these differences do not imply, in the compositional data analysis approach, a significant change in covariance structure. Under compositional data analysis, covariance-based analysis should give similar results regardless of the size fraction adopted, while distance-based ones may give partly different results. The 106 m size fraction, in which both large and small-sized species are sufficiently represented, is proposed for adoption in future studies focussing on palaeoclimatic reconstruction, This research was supported by the Ministerio de Economia y Competividad under the project METRICS Ref. MTM2012-33236; by the Agencia de Gestio d'Ajuts Universitaris i de Recerca of the Generalitat de Catalunya under the project Ref. 2009SGR424. The authors would like to thank the anonymous reviewers for their valuable comments and suggestions to improve the quality of the paper.
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Calorific value and compositional ultimate analysis with a case study of a Texas lignite

  • Olea, Ricardo A.
  • Luppens, James A.
  • Egozcue, Juan José
  • Pawlowsky-Glahn, Vera
Measurements to determine coal quality as fuel include proximate analysis, ultimate analysis and calorific value. The latter is an attribute taking non-negative real values, so a simple transformation is sufficient for its spatial modeling applying geostatistics. The analyses, however, involve proportions that follow the properties of compositional data, thus requiring special preprocessing for an adequate modeling already described in a previous publication for the case of proximate analysis data.11Olea, R.A., Luppens, J.A., 2015. Mapping of coal quality using stochastic simulation and isometric logratio transformation with an application to a Texas lignite. International Journal of Coal Geology, 152, 80-93. Here we model the results of calorific value and ultimate analysis. We propose to use two different binary partitions, one per analysis, map the corresponding isometric logratio transformations, and backtransform the results. The methodology is illustrated using the same coal bed in the previous paper modeling proximate analysis data. Results are summarized using probability maps that, in the case of this deposit, show a prominent channel crossing the deposit and separating the best quality coal from that of lower quality, J.J. Egozcue and V. Pawlowsky-Glahn have been supported by the Ministerio de Economía y Competividad (Spain) under the project “CODA-RETOS” (Ref. MTM2015-65016-C2-1-R) and the project “METRICS” (Ref.MTM2012-33236); and by the AGAUR of the Generalitat de Catalunya under the project “COSDA” (Ref: 2014SGR551)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Methods to investigate the geochemistry of groundwaters with values for nitrogen compounds below the detection limit

  • Buccianti, Antonella
  • Nisi, Barbara
  • Martín Fernández, Josep Antoni
  • Palarea Albaladejo, Javier
Groundwaters, like other natural waters, are solutions of a variety of substances in the solvent water. Human impact upon groundwater systems has created many environmental problems so that hydrogeochemical studies related to pollution have become very important. In this context high abundance of nitrogen species, particularly nitrates, can cause adverse health effects, their main sources being fertilizer, animal or human waste, natural soil organic matter, nitrogen fixation and rain. Graphical representations of hydrogeochemical data (for example molar ratio diagrams, stochiometric diagrams, triangular diagrams, mixing diagrams) pose considerable problems if statistical evaluations have to be performed. In fact, since hydrochemical data are compositional (proportional data), their sample space is the simplex, a constrained space where the application of the standard geometrical approach may provide misleading information. In this paper classical binary diagrams used to investigate the whole chemistry of water as well as the behaviour of nitrogen species were substituted by new equivalent graphs coherent with the properties of compositional data, thus opening new perspectives in the evaluation of geochemical processes affecting water resources. Moreover, since nitrogen species are often affected by the presence of numerous data below the detection limit, their role was investigated by considering different imputation methods, and evaluating their performance in the sub-composition NH4+,NO2-, NO3-, NO3-. The first phase of the analysis did not consider the presence of data below the detection limit but subsequently, step by step, the possibility to recuperate this type of information was explored. The opportunity to recover information about data below the detection limit from the variance-covariance structure of the whole composition (Ca-2(+), Mg2+, K+, Na+, HCO3-, Cl- and SO42- and available nitrogen data) was also investigated by considering a multivariate framework. The approaches, coherent with compositional data analysis theory, pointed out interesting aspects of 1) the relationships in the sub-composition of NH4+, NO2-, and NO3-, giving us a different perception of the background level of these species as well as of their source; 2) the effects on the variance-covariance structure of the whole composition (main anions and cations and nitrogen species); and 3) the loss of information about the covariance between chemical variables (coherence of geochemical behaviour), which is able to affect the understanding of geochemical systems, This research has been supported by the University of Florence (ex-60% 2011/2012 funds), Geobasi Project (Tuscany Region, Italy), Scottish Government's Rural and Environment Science and Analytical Services (RESAS) and "METRICS" (Ref. MTM2012-33236) project of the Spanish Ministry of Economy and Competitiveness
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

It's all relative: analyzing microbiome data as compositions

  • Gloor, Gregory B.
  • Wu, J.R.
  • Pawlowsky-Glahn, Vera
  • Egozcue, Juan José
Purpose: The ability to properly analyze and interpret large microbiome datasets has lagged behind our ability to acquire such datasets from environmental or clinical samples. Sequencing instruments impose a structure on these data: the natural sample space of a 16S rRNA gene sequencing dataset is a simplex, which is a part of real space that is restricted to non-negative values with a constant sum. Such data are compositional, and should be analyzed using compositionally appropriate tools and approaches. However, the majority of the tools for 16S rRNA gene sequencing analysis assume these data are unrestricted. Methods: We show that existing tools for compositional data (CoDa) analysis can be readily adapted to analyze high throughput sequencing datasets. Results: The Human Microbiome Project tongue vs. buccal mucosa dataset shows how the CoDa approach can address the major elements of microbiome analysis. Reanalysis of a publicly available autism microbiome dataset shows that the CoDa approach in concert with multiple hypothesis test corrections prevent false positive identifications. Conclusions: The CoDa approach is readily scalable to microbiome-sized analyses. We provide example code, and make recommendations to improve the analysis and reporting of microbiome datasets, Work in G.B. Gloor’s lab has been supported by a Discovery Grant from the National Science and Engineering Research Council of Canada. J.R. Wu was supported by a CIHR grant to Dr. J. Allard and GBG. Drs J.J. Egozcue and V. Pawlowsky-Glahn have been supported by the Spanish Ministry of Economy and Competitiveness under the project METRICS (Ref. MTM2012-33236); and by the Agència de Gestió d’Ajuts Universitaris i de Recerca of the Generalitat de Catalunya under the project COSDA (Ref: 2014SGR551).
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Log-ratio methods in mixture models for compositional data sets

  • Comas Cufí, Marc
  • Martín Fernández, Josep Antoni
  • Mateu i Figueras, Glòria
When traditional methods are applied to compositional data misleading and incoherent results could be obtained. Finite mixtures of multivariate distributions are becoming increasingly important nowadays. In this paper, traditional strategies to fit a mixture model into compositional data sets are revisited and the major difficulties are detailed. A new proposal using a mixture of distributions defined on orthonormal log-ratio coordinates is introduced. A real data set analysis is presented to illustrate and compare the different methodologies, This research was supported by the Ministerio de Economía y Competividad through the projects “METRICS” and “CoDa-RETOS” (MTM2012-33236; MTM2015-65016- C2-1-R: MINECO/FEDER,UE) and the Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR: 2014SGR551)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Spatial analysis of compositional data: A historical review

  • Pawlowsky-Glahn, Vera
  • Egozcue, Juan José
Like the statistical analysis of compositional data in general, spatial analysis of compositional data requires specific tools. A historical overview of their development is presented in three steps: (a) the recognition of the problem, known as spurious spatial covariance, (b) first attempts to use the logratio approach, and (c) the application of the principle of working in coordinates using isometric logratio representations. Also mentioned are the use of matrix-valued variation-variograms as a tool to model crossvariograms, and the simplicial approach to indicator kriging, that solves inconsistencies in the standard approach to indicator kriging, This research has been supported by the SpanishMinistry of Education and Science under project ‘METRICS’ (Ref. MTM2012-33236); and from the Agència de Gestió d'Ajuts Universitaris i de Recerca of the Generalitat de Catalunya under the project Ref. 2009SGR424
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Individual T-2 Control Chart for Compositional Data

  • Vives Mestres, Marina
  • Daunis-i-Estadella, Pepus
  • Martín Fernández, Josep Antoni
The usual Hotel ling T-2 control chart is not appropriate for monitoring processes where the quality characteristic is a mixture. The composition of mixtures are vectors of positive elements that represent parts of a whole, to which standard multivariate techniques are not appropriate due to their restricted sample space. There are many applications where a mixture is monitored against time, such as in the chemical industry, product composition, impurity profile, or gas components analysis. In this paper, a multivariate control chart for individual compositional observations based on the T-2 statistic is proposed and compared with the typical one in terms of average run length. We show how results are more consistent with compositional data nature and illustrate implementation in a real-world example, This work has been partially financed by the Agencia de Gestio d'Ajuts Universitaris i de Recerca (Ref: 2009SGR424) and by the Ministerio de Ciencia e Innovacion (Ref: MTM2012-33236 and Ref: MTM2009-13272). The authors wish to thank the Universitat de Girona for the grant given to Marina Vives-Mestres during her doctoral studies
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Compositional methods for estimating elemental concentrations below the limit of detection in practice using R

  • Palarea Albaladejo, Javier
  • Martín Fernández, Josep Antoni
  • Buccianti, Antonella
Geochemical studies typically generate data sets including labels for values below certain limits of detection of the analytical techniques involved. These unobserved values hamper subsequent data analysis, and a number of procedures with different levels of sophistication have been introduced in order to deal with them. Unfortunately, those most commonly used in practice are those most prone to introduce bias and lead to poor estimates of the statistics of interest. We approach this topic in the context of geochemical compositional data, that is, data representing parts of a whole. Although the general statistical literature has provided well-based methods for censored data, which includes the case of nondetects, they have ignored the special nature of compositional samples. In this paper we review a number of methods introduced in the recent years to deal with values below the detection limit while meeting compositional principles. We provide guidance and computer routines to easily apply them in practice using the popular R statistical computing language. © 2013 Elsevier B.V., This research has been supported by the Scottish Government Rural and Environment Science and Analytical Services (RESAS) and the project "METRICS" (Ref. MTM2012-33236), Spanish Ministry of Economy and Competitiveness
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Compositional data analysis (CoDA) as a tool to study the (paleo)ecology of coccolithophores from coastal-neritic settings off central Portugal

  • Guerreiro, Catarina
  • Cachão, Mário
  • Pawlowsky-Glahn, Vera
  • Oliveira, Anabela
  • Rodrigues, Aurora
Whereas using the species percentages is the standard analytical procedure used to infer species ecological preferences, independently of taphonomical effects, the closure problem associated with closed number systems and subsequent inconsistency of determining percentages may lead to spurious correlations, biased statistical analysis and misleading interpretations. To avoid these problems, we applied Compositional Data Analysis (CoDA) to investigate the (paleo)ecological preferences and spatial distribution of coccolith assemblages preserved in seafloor sediments, using as a case-study the central Portuguese submarine canyons and adjacent shelf-slope areas. Results from using the isometric log-ratio (ilr) approach from CoDA are compared with results from using classical analytical methods, and further discussed. While providing scale invariance and subcompositional coherence, CoDA is revealed to be a consistent statistical tool to infer the (paleo)ecological preferences of coccolithophores, corroborating earlier work based on from percentage determinations. Results of this study clearly confirmed the coastal-neritic distribution of coccoliths from Gephyrocapsa oceanica, Helicosphaera carteri and Coronosphaera mediterranea, whereas coccoliths from Calcidiscus leptoporus, Umbilicosphaera sibogae, Umbellosphaera irregularis and Rhabdosphaera spp. more typically occur offshore. Differences between canyons and adjacent shelf and slope areas were also confirmed, namely the (a) greater importance of the coastal-neritic assemblage possibly resulting from local and persistent nutrient pumping in these areas, and (b) stronger mixing of coccoliths from coastal and oceanic species in upper canyon reaches, resulting from the focused coastward advection of more oceanic water masses along their axes. Unlike the results from both ilr-coordinates and the percentage approaches, both coccolith concentrations and fluxes showed that spatial trends in which the species ecological inter-relationships appear to be masked by taphonomical phenomena, especially towards the coast and in the canyons, suggesting the two latter approaches are not suitable to perform (paleo)ecological inferences in more dynamic coastal-neritic settings. Our study suggests that the (paleo)ecological signal preserved in the studied sediment samples is persistent enough to be revealed by both CoDA and percentages. Yet, given that CoDA provides the only statistical solution to coherently draw (paleo)ecological interpretations from compositional data, our recommendation is that CoDA should always be used to test and validate any ecological signals obtained from percentage distributions, Data for this study were collected in the framework of the EU-funded EUROSTRATAFORM (contract EVK3-CT-2002-00079) and HERMES EU-funded projects (contract GOCE-CT-2005-511234), "Lead in Canyons" and "Pacemaker" projects funded by the Netherlands Organization for Scientific Research, and by the Portuguese project Cd Tox-CoN (FCTPTDC/MAR/102800/2008). Multicores were collected during RV Pelagia cruises which were funded by the Netherlands Organisation for Scientific Research. Samples were prepared and analysed at NANOLAB, CEGUL. A special thanks to Henko De Stigter for supplying sample material and for textual improvements and to Marc Comas for his help with CODAPACK. The first author benefited from a PhD grant (FCT-SFRH/BD/41330/2007), a post-doctoral fellowship from the University of Lisbon in the context of "Programa Estrategico do Instituto Dom Luiz" (PEST-OE/CTE/LA0019/2013-2014), funded by national funds of FCT/MEC (PIDDAC), and currently benefits from a Marie Curie post-doctoral fellowship attributed by the BREMEN-TRAC COFUND program, in the context of the 7th Framework Programme of the European Commission. V. Pawlowsky-Glahn is supported by the Spanish Ministry of Economy and Competitiveness through the project METRICS Ref. MTM2012-33236. The manuscript benefited from a detailed and helpful revision from Dr. Jasper Knight and two anonymous reviewers, for which the authors are most grateful
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Bayesian-multiplicative treatment of count zeros in compositional data sets

  • Martín Fernández, Josep Antoni
  • Hron, Karel
  • Templ, Matthias
  • Filzmoser, Peter
  • Palarea Albaladejo, Javier
Compositional count data are discrete vectors representing the numbers of outcomes falling into any of several mutually exclusive categories. Compositional techniques based on the log-ratio methodology are appropriate in those cases where the total sum of the vector elements is not of interest. Such compositional count data sets can contain zero values which are often the result of insufficiently large samples. That is, they refer to unobserved positive values that may have been observed with a larger number of trials or with a different sampling design. Because the log-ratio transformations require data with positive values, any statistical analysis of count compositions must be preceded by a proper replacement of the zeros. A Bayesian-multiplicative treatment has been proposed for addressing this count zero problem in several case studies. This treatment involves the Dirichlet prior distribution as the conjugate distribution of the multinomial distribution and a multiplicative modification of the non-zero values. Different parameterizations of the prior distribution provide different zero replacement results, whose coherence with the vector space structure of the simplex is stated. Their performance is evaluated from both the theoretical and the computational point of view, This research was supported by the Ministerio de Economia y Competividad under the project 'METRICS' Ref. MTM2012-33236, by the Agencia de Gestio d'Ajuts Universitaris i de Recerca of the Generalitat de Catalunya under the project Ref: 2009SGR424, and by the Scottish Government's Rural and Environment Science and Analytical Services Division (RESAS). The authors also gratefully acknowledge the support by the Operational Program Education for Competitiveness-European Social Fund (project CZ.1.07/2.3.00/20.0170 of the Ministry of Education, Youth and Sports of the Czech Republic)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Polymer incremental sheet forming process: temperature analysis using response surface methodology

  • Bagudanch Frigolé, Isabel
  • Garcia-Romeu, Maria Luisa
  • Vives Mestres, Marina
  • Sabater i Armengou, Marc
To reduce costs associated with the manufacturing of customized products, several innovative forming processes have been developed. Incremental sheet forming (ISF) is one of these new technologies, becoming, in the past decade, more interesting for the academic and industrial community. The influence of main process parameters, namely, tool diameter, spindle speed, feed rate, and step down, is studied in depth in this paper. The maximum temperature achieved during the forming process of a truncated pyramid frustum with a circular generatrix using three nonbiocompatible and two biocompatible polymer materials is measured. Box–Behnken design of experiments and the response surface methodology have been utilized to statistically analyze the results and to provide models able to predict the maximum temperatures, This research has received funding from the Spanish Ministry of Education (DPI2012-36042), the Spanish Ministry of Economy and Competitiveness (MTM2012-33236 and MTM2015-65016-C2-1-R), the University of Girona (MPCUdG2016/036) and the Catalan Agency for Management of University and Research Grants (2014 SGR551)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

The peril of proportions: robust niche indices for categorical data

  • Pierotti, Michele E. R.
  • Martín Fernández, Josep Antoni
  • Barceló i Vidal, Carles
Indices of niche breadth and niche overlap for categorical data are typically expressed in terms of proportions of resources use. These are unit-sum constrained data; hence, direct application of standard general linear modelling methods to such indices can lead to spurious correlations and misleading inference. To overcome these limitations, we introduce a compositional data analysis (CoDA) approach and derive compositional expressions of niche breadth, niche overlap and specialization. Compositional data analysis is specifically devoted to the analysis of vectors of proportions (i.e. compositions) and represents the appropriate framework for the study of sets of data with unit-sum constraint as those typically used in the calculation of niche indices. We show that compositional indices exhibit suitable statistical properties that make them flexible and robust, allowing downstream application of the full toolbox of multivariate analysis techniques to these estimators, a possibility not available with classical indices. In addition, we find that when characterizing niche breadth, niche overlap and specialization in terms of vectors of proportions, these concepts are naturally integrated in a coherent unifying framework. When data are categorical, we recommend the use of compositional indices for the statistical analysis of specialization metrics, niche breadth and niche overlap. We believe that the unified framework emerging from our compositional approach to niche metrics will allow a more thorough understanding of specialization at multiple levels of biological organization and provide novel insights in complex phenomena such as invasions and niche shifts, This research was supported by the Ministerio de Econom ıa y Competitividad under the project ‘METRICS’ (Ref. MTM2012-33236), and by the AGAUR of the Generalitat de Catalunya under the project ‘COSDA’ (Ref: 2014SGR551)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Proportionality: A Valid Alternative to Correlation for Relative Data

  • Lovell, David
  • Pawlowsky-Glahn, Vera
  • Egozcue, Juan José
  • Marguerat, Samuel
  • Bähler, Jürg
In the life sciences, many measurement methods yield only the relative abundances of different components in a sample. With such relative-or compositional-data, differential expression needs careful interpretation, and correlation-a statistical workhorse for analyzing pairwise relationships-is an inappropriate measure of association. Using yeast gene expression data we show how correlation can be misleading and present proportionality as a valid alternative for relative data. We show how the strength of proportionality between two variables can be meaningfully and interpretably described by a new statistic. which can be used instead of correlation as the basis of familiar analyses and visualisation methods, including co-expression networks and clustered heatmaps. While the main aim of this study is to present proportionality as a means to analyse relative data, it also raises intriguing questions about the molecular mechanisms underlying the proportional regulation of a range of yeast genes, DL was funded by the Commonwealth Scientific and Industrial Research Organisation (www.csiro.au). VPG was funded by the Spanish Ministry of Education, Culture and Sports under a Salvador de Madariaga grant (Ref. PR2011-0290) and by the Spanish Ministry of Economy and Competitiveness under the project METRICS Ref. MTM2012-33236. JJE was funded by the Agencia de Gestio d'Ajuts Universitaris i de Recerca of the Generalitat de Catalunya under project Ref: 2009SGR424. SM was funded by the UK Medical Research Council. JB was funded by a Wellcome Trust Senior Investigator Award (grant #095598/Z/11/Z)