A5101465028- X-ray Diffraction in Crystallography
- Crystallization and Solubility Studies
- Machine Learning and Data Classification
- Gene expression and cancer classification
- Bayesian Modeling and Causal Inference
- Metaheuristic Optimization Algorithms Research
- Machine Learning in Bioinformatics
- Face and Expression Recognition
- Anomaly Detection Techniques and Applications
- Bayesian Methods and Mixture Models
- Imbalanced Data Classification Techniques
- Cancer Genomics and Diagnostics
- Algorithms and Data Compression
- Genomics and Phylogenetic Studies
- Bioinformatics and Genomic Networks
- Machine Learning and Algorithms
- Advanced Multi-Objective Optimization Algorithms
- Crystallography and molecular interactions
- Mobile Ad Hoc Networks
- Evolutionary Algorithms and Applications
- Data Mining Algorithms and Applications
- Genome Rearrangement Algorithms
- VLSI and FPGA Design Techniques
- Evolution and Genetic Dynamics
- Pancreatic function and diabetes
University of the Basque Country
2013-2023
Centro Tecnolóxico de Telecomunicacións de Galicia
2016
University of Havana
2006
Comparing the results obtained by two or more algorithms in a set of problems is central task areas such as machine learning optimization.Drawing conclusions from these comparisons may require use statistical tools hypothesis testing.There are some interesting papers that cover this topic.In manuscript we present scmamp, an R package aimed at being tool simplifies whole process analyzing when comparing algorithms, loading data to production plots and tables.
Differences in gene expression patterns have been documented not only Multiple Sclerosis patients versus healthy controls but also the relapse of disease. Recently a new modulator has identified: microRNA or miRNA. The aim this work is to analyze possible role miRNAs multiple sclerosis, focusing on stage. We analyzed 364 PBMC obtained from sclerosis status, remission status and controls. with significantly different were validated an independent set samples. In order determine effect miRNAs,...
In this paper we present the R package PerMallows, which is a complete toolbox to work with permutations, distances and some of most popular probability models for permutations: Mallows Generalized models. The model an exponential location model, considered as analogous Gaussian distribution. It based on definition distance between permutations. its best-known extension. includes functions making inference, sampling learning such distributions. in PerMallows are Kendall's τ , Cayley, Hamming Ulam.
The statistical assessment of the empirical comparison algorithms is an essential step in heuristic optimization. Classically, researchers have relied on use tests. However, recently, concerns about their arisen and, many fields, other (Bayesian) alternatives are being considered. For a proper analysis, different aspects should be In this work we focus question: what probability given algorithm best? To tackle question, propose Bayesian analysis based Plackett-Luce model over rankings that...
The most commonly used statistics in Evolutionary Computation (EC) are of the Wilcoxon-Mann-Whitney-test type, its either paired or non-paired version. However, using such for drawing performance comparisons has several known drawbacks. At same time, Bayesian inference analysis is an emerging statistical tool, which potential to become a promising complement perspectives offered by aforementioned p-value type test. This work exhibits practical use typical EC setting, where algorithms be...
The development of techniques for oncogenomic analyses such as array comparative genomic hybridization, messenger RNA expression arrays and mutational screens have come to the fore in modern cancer research. Studies utilizing these are able highlight panels genes that altered cancer. However, candidate must then be scrutinized reveal whether they contribute oncogenesis or coincidental non-causative. We present a computational method prioritization (i) proto-oncogenes (ii) tumour suppressor...
AbstractA biocompatible emulsification method for microencapsulation of live cells and enzymes within a calcium alginate matrix applied to Bacillus Calmette-Gukrin (BCG) has been developed. Small-diameter beads (microcapsules) were formed via internal gelation an solution emulsified vegetable oil., Five different oils (sesamesweet almondperhydrosqualenecamomile jojoba) used., The rheological analysis the showed Newtonian behaviourwith viscosities = 30.037.751.259.3 67.1 mPa.s...
Luminal B breast tumors have aggressive clinical and biological features, constitute the most heterogeneous molecular subtype, both clinically molecularly. Unfortunately, immunohistochemistry correlate of luminal subtype remains still imprecise, it has now become paramount importance to define a classification scheme capable segregating into meaningful subgroups that may be used guide patient management. With aim unraveling DNA methylation profiles subtypes currently being in setting, we...
<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> Microarray-based global gene expression profiling, with the use of sophisticated statistical algorithms is providing new insights into pathogenesis autoimmune diseases. We have applied a novel technique for selection based on machine learning approaches to analyze microarray data gathered from patients systemic lupus erythematosus (SLE) and primary antiphospholipid syndrome (PAPS), two diseases...
The Mallows and Generalized Models are two of the most popular probability models for distributions on permutations. In this paper, we consider both under Hamming distance. This can be seen as matchings instead rankings. These cannot factorized, which contrasts with MM GMM Kendall's-$\tau$ Cayley distances. order to overcome computational issues that involve, introduce a novel method computing partition function. By adapting compute expectation, joint conditional probabilities. All these...
In the field of optimization and machine learning, statistical assessment results has played a key role in conducting algorithmic performance comparisons. Classically, null hypothesis tests have been used. However, recently, alternatives based on Bayesian statistics shown great potential complex scenarios, especially when quantifying uncertainty comparison. this work, we delve deep into experimental by proposing framework for analysis several algorithms problems/instances. To end, are...