A5031078665- Geochemistry and Geologic Mapping
- Soil Geostatistics and Mapping
- Hydrocarbon exploration and reservoir analysis
- Geological and Geochemical Analysis
- Mineral Processing and Grinding
- Heavy metals in environment
- Remote-Sensing Image Classification
- earthquake and tectonic studies
- Hydrology and Drought Analysis
- Seismic Performance and Analysis
- Ocean Waves and Remote Sensing
- Structural Health Monitoring Techniques
- Bayesian Methods and Mixture Models
- Climate variability and models
- Atmospheric and Environmental Gas Dynamics
- Advanced Statistical Methods and Models
- Financial Risk and Volatility Modeling
- Gut microbiota and health
- Oceanographic and Atmospheric Processes
- Landslides and related hazards
- Water Quality and Pollution Assessment
- Probabilistic and Robust Engineering Design
- Seismic Waves and Analysis
- Statistical Methods and Inference
- Rough Sets and Fuzzy Logic
Universitat Politècnica de Catalunya
2015-2024
Research Center for Natural Resources, Environment and Society
2021
Polytechnic Institute of Castelo Branco
2021
University of Évora
2021
Universitat de Barcelona
2008
Polytechnic University of Puerto Rico
1991-1992
Numerical Method (China)
1988
Abstract Compositional data are those which contain only relative information. They parts of some whole. In most cases they recorded as closed data, i.e. summing to a constant, such 100% — whole-rock geochemical being classic examples. have important and particular properties that preclude the application standard statistical techniques on in raw form. Standard designed be used with free range from − ∞ + ∞. always positive 0 100, or any other when given If one component increases, others...
In the life sciences, many measurement methods yield only 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 association. Using yeast gene data we show how correlation can be misleading present proportionality as valid alternative data. We strength between two variables meaningfully...
We propose a new algorithm for the identification of microbial signatures. These signatures can be used diagnosis, prognosis, or prediction therapeutic response based on an individual’s specific microbiota.
Abstract The main features of the Aitchison geometry simplex D parts are reviewed. Compositions positive vectors in which relevant information is contained ratios between their components or parts. They can be represented by closing them to a constant sum, e.g. percentages, per million. Perturbation and powering respectively an internal operation, playing role external product real numbers scalars. These operations impose structure ( − 1)-dimensional vector space An inner product, norm...
Within the special geometry of simplex, sample space compositional data, orthonormal coordinates allow application any multivariate statistical approach. The search for meaningful has suggested balances (between two groups parts)—based on a sequential binary partition D-part composition—and representation in form CoDa-dendrogram. Projected samples are represented dendrogram-like graph showing: (a) way grouping parts; (b) explanatory role subcompositions generated process; (c) decomposition...
Summary A Bayes linear space is a of equivalence classes proportional σ‐finite measures, including probability measures. Measures are identified with their density functions. Addition given by Bayes' rule and substraction Radon–Nikodym derivatives. The present contribution shows the subspace square‐log‐integrable densities to be Hilbert space, which can include infinite measures on whole real line or discrete It extends ideas from finite support towards spaces general measure spaces. also...
The most suitable method for estimation of size diversity is investigated. Size computed on the basis Shannon expression adapted continuous variables, such as size. It takes form an integral involving probability density function (pdf) individuals. Different approaches pdf are compared: parametric methods, assuming that data come from a determinate family pdfs, and nonparametric where estimated using some kind local evaluation. Exponential, generalized Pareto, normal, log‐normal...
Tissue analysis is commonly used in ecology and agronomy to portray plant nutrient signatures. Nutrient concentration data, or ionomes, belongs the compositional data class, i.e. multivariate that are proportions of some whole, hence carrying important numerical properties. Statistics computed across raw ordinary log-transformed intrinsically biased, possibly leading wrong inferences. Our objective was present a sound robust approach based on novel balance concept classify ionomes. We...
Fecal microbiota transplantation (FMT) has been recently approved by FDA for the treatment of refractory recurrent clostridial colitis (rCDI). Success FTM in rCDI led to a number studies investigating effectiveness its application other gastrointestinal diseases. However, majority effects FMT were evaluated on patients with initially altered microbiota. The aim our study was estimate gut composition healthy volunteers and monitor long-term outcomes.We have performed combined analysis three...
Compositional data analysis usually deals with relative information between parts where the total (abundances, mass, amount, etc.) is unknown or uninformative. This article addresses question of what to do when known and interest. Tools used in this case are reviewed analysed, particular relationship positive orthant D-dimensional real space, product space line times D-part simplex, their Euclidean structures. The first alternative corresponds taking logarithms on each component, second one...