Strategic Research Soft Computing (SOCO)

Soft Computing (SOCO)

The main objectives of the SOCO group are twofold:

One the one, we aim to make progress in the state-of-the-art of soft-computing methodologies, as well as to investigate their possible hybridisation in order to improve their robustness and efficacy. The SOCO group is currently working on soft-computing methods including, among others, the following:

  • Artificial neural networks (feed-forward, recurrent, heterogeneous)
  • Support vector machines
  • Deep Learning
  • Fuzzy Inductive Reasoning
  • Fuzzy control
  • Genetic algorithms and evolutionary strategies
  • Unsupervised probabilistic models
  • Pattern recognition and computer vision
  • Feature selection and dimensionality reduction

On the other hand, we aim to apply these methods to real-world problems. The group has long-term experience in the following areas of application:

  • Medical (human central nervous system, cancer prediction and diagnosis, cognitive neuroscience, major depression, Alzeheimer, etc.)
  • Pharmacological (G-Protein Coupled Receptors analysis)
  • Biological (growth of white shrimp, wastewater treatment plants)
  • Business (church analysis, market segmentation)
  • Environmental (seismic risk estimation, ozone and air contaminants)
  • Music (Modelling high-level musical features)

Main projects

  • 1.

    KAPPA-AIM: Knowledge Acquisition in Pharmacoproteomics using Advanced Artificial Intelligence Methods

    Period: 2013-01-01 to 2015-12-31 More than 50% of drugs target only four key protein families, from which almost a 30% correspond to the G-Protein Coupled Receptors (GPCR) superfamily. The central objective of this project was dual: on one side, the design of adequate artificial intelligence-based methods for the analysis of GPCR sequential data. On the other side, and given that this research has the ultimate goal of being useful in helping drug design and the understanding of the molecular processes involved, we aimed to apply the developed methods in relevant pharmacoproteomic problems such as GPCR subtyping, receptor heteromerization and deorphanization, and protein alignment-free analysis

  • 2.

    AIDTumour: Herramientas Basadas en Métodos de Inteligencia Artificial para el Apoyo a la Decisión en Oncología

    Period: 2010-01-01 to 2012-12-31 The core of this project was the design and development of artificial intelligence-related tools for decision support in the clinical diagnosis of human cancer pathologies, with a focus on brain tumours. The tools developed in this project made use of advanced techniques from the fields of Machine Learning, Computational Intelligence, Pattern Recognition and Soft Computing, to solve problems of classification, clustering, feature selection and extraction, time series analysis and data visualization.

  • 3.

    (MADep) Monitoratge i Assistència per a la Depressió

    Period: 2009-06-05 to 2011-06-05 Major depression is one of the most common psychiatric disorders present in the general population, with a lifetime prevalence of 12.8%, and 12-month prevalence range between 3.6% and 9.1% in Europe. In this project, a major depression monitoring system was developed to facilitate a more frequent follow-up of patient progress during their home stay after being diagnosed with depression by a psychiatrist.



Bioinformatics expertise:

Group Leader:

Maria Angela Nebot Castells


Alfredo Vellido Alcacena

Bioinformatics services offered

  • The SOCO research group

    The SOCO research group provides expertise on data analytics for bioinformatics and biomedical problems. We are in a key position between pure technical and IT proficiency and biomedical/bioinformatics expertise that allows us to generate bespoke solutions in these fields. Our particular expertise on state-of-the-art artificial intelligence and machine learning provides us with a powerful set of tools to address the everincreasing data-related challenges found in proteomics, genomics and evidence-based medicine