Pasar al contenido principal

INTELIGENCIA COMPUTACIONAL PARA EL DESCUBRIMIENTO DE CONOCIMIENTO A PARTIR DE G PROTEIN-COUPLED RECEPTORS

TIN2016-79576-R

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Convocatoria Proyectos I+D+i Retos de Investigación
Año convocatoria 2016
Plan Estatal 2013-2016 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 7
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Topological sequence segments discriminate between class C GPCR subtypes

  • König, Caroline|||0000-0002-7543-8686
  • Alquézar Mancho, René|||0000-0002-6420-0517
  • Vellido Alcacena, Alfredo|||0000-0002-9843-1911
  • Giraldo Arjonilla, Jesús
G protein-coupled receptors are eukaryotic cell membrane proteins with a key role as extracellular signal transmitters. While GPCRs embrace a wide and heterogeneous super-family of proteins, our interest in this study is in its Class C, of great relevance to pharmacology. The scarcity of knowledge about their full 3-D crystal structure makes the use of their primary amino acid sequences important for analysis. In this paper, we systematically analyze whether segments of the receptor sequences are able to discriminate between the different class C GPCR subtypes according to their topological location on the extracellular, transmembrane or intracellular domain. For this, we build on previous research that showed that the use of the extracellular N-terminus domain on its own for this classification task did only entail a minor decrease in subtype discrimination when compared to the complete sequence. We use Support Vector Machine-based classification models to assess the subtype discriminating power of the topological segments., This research was partially funded by the Spanish MINECO TIN2016-79576-R and SAF2014-58396-R project., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Societal issues in machine learning: when learning from data is not enough

  • Bacciu, Davide
  • Biggio, Battista
  • Lisboa, Paulo J G
  • Martin Guerrero, Jose David
  • Oneto, Luca
  • Vellido Alcacena, Alfredo
It has been argued that Artificial Intelligence (AI) is experiencing a fast process of commodification. Such characterization is on the interest of big IT companies, but it correctly reflects the current industrialization of AI. This phenomenon means that AI systems and products are reaching the society at large and, therefore, that societal issues related to the use of AI and Machine Learning (ML) cannot be ignored any longer. Designing ML models from this human-centered perspective means incorporating human-relevant requirements such as safety, fairness, privacy, and interpretability, but also considering broad societal issues such as ethics and legislation. These are essential aspects to foster the acceptance of ML-based technologies, as well as to ensure compliance with an evolving legislation concerning the impact of digital technologies on ethically and privacy sensitive matters. The ESANN special session for which this tutorial acts as an introduction aims to showcase the state of the art on these increasingly relevant topics among ML theoreticians and practitioners. For this purpose, we welcomed both solid contributions and preliminary relevant results showing the potential, the limitations and the challenges of new ideas, as well as refinements, or hybridizations among the different fields of research, ML and related approaches in facing real-world problems involving societal issues., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Interpreting response to TMZ therapy in murine GL261 glioblastoma by combining Radiomics, Convex-NMF and feature selection in MRI/MRSI data analysis

  • Nuñez Vivero, Luis Miguel
  • Julia Sape, Margarida
  • Romero Merino, Enrique
  • Arus Caraltó, Carles
  • Vellido Alcacena, Alfredo
  • Candiota Silveira, Ana Paula
Machine learning (ML) methods have shown great potential for the analysis of data involved in medical decisions. However, for these methods to be incorpored in the medical pipeline, they must be made interpretable not only to the data analyst, but also to the medical expert. In this work, we have applied a combination of feature transformation, selection and classification using ML and statistical methods to differentiate between control (untreated) and Temozolomide (TMZ)-treated tumour tissue from a glioblastoma (brain tumour) murine model. As input, we have used T2 weighted magnetic resonance images (MRI) and spectroscopic imaging (MRSI). Radiomics features have been extracted from the MRI dataset, while convex Non-negative Matrix Factorization (Convex-NMF) was used to extract sources from the MRSI dataset. Exhaustive feature selection has revealed parsimonious feature subsets that facilitate the expert interpretation of results while retaining a high discriminatory ability., L.M.N., A.V., C.A. and A.P.C. received EC funding from the ATTRACT project, under Grant Agreement 777222. A.V. and E.R. acknowledge funding from Spanish MINECO TIN2016- 79576-R research project., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

K nearest neighbour optimal selection in fuzzy inductive reasoning for smart grid applications

  • Jurado Gómez, Sergio
  • Nebot Castells, M. Àngela
  • Múgica Álvarez, Francisco
Energy recasting has been an area of great interest in the last years. It unlocks, not only the Smart Grid's potential with load balancing but also new business models and added value services. To achieve an accurate, robust and fast prediction, model's parametrization is key and becomes a bottleneck in the value-chain. In this article, we present an improved version of Flexible Fuzzy Inductive Reasoning (Flexible FIR) that selects the most optimal number of nearest neighbours during FIR prediction phase, called K nearest neighbour Optimal Selection (KOS). To this end, a real smart grid forecasting application, i.e. electricity load forecasting, has been chosen in this study. The results show that the best forecasting accuracy, on average, is when the KOS is used on Flexible FIR. While with KOS the optimal parameter k is found online, without it is not, which increases the computational time., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Topological sequence segments discriminate between class C GPCR subtypes

  • König, Caroline
  • Alquézar Mancho, René
  • Vellido Alcacena, Alfredo
  • Giraldo Arjonilla, Jesús
G protein-coupled receptors are eukaryotic cell membrane proteins with a key role as extracellular signal transmitters. While GPCRs embrace a wide and heterogeneous super-family of proteins, our interest in this study is in its Class C, of great relevance to pharmacology. The scarcity of knowledge about their full 3-D crystal structure makes the use of their primary amino acid sequences important for analysis. In this paper, we systematically analyze whether segments of the receptor sequences are able to discriminate between the different class C GPCR subtypes according to their topological location on the extracellular, transmembrane or intracellular domain. For this, we build on previous research that showed that the use of the extracellular N-terminus domain on its own for this classification task did only entail a minor decrease in subtype discrimination when compared to the complete sequence. We use Support Vector Machine-based classification models to assess the subtype discriminating power of the topological segments., This research was partially funded by the Spanish MINECO TIN2016-79576-R and SAF2014-58396-R project., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Charting perceptual spaces with fuzzy rules

  • Paz Ortiz, Alejandro Iván
  • Nebot Castells, M. Àngela
  • Romero Merino, Enrique
  • Múgica Álvarez, Francisco
Algorithmic music nowadays performs domain specific tasks for which classical algorithms do not offer optimal solutions or require user's expertise. Among these tasks is the extraction of models from data that offer an understanding of the underlying behavior, providing a quick and easy to use way to explore the data for first (sometimes on-the-fly) insights. Learning rules from examples is an approach often used to achieve this goal. However, together with the aforementioned requirements algorithmic composition needs to create new material so that it is perceived as consistent with the material of the data. In addition, the input data sets are usually small because the human is the bottleneck when generating them. In this contribution we present a fuzzy rule induction algorithm focused on generalizing a set of data, complying with the previous requirements, that offers good results for small data sets. For its evaluation -in a field where there are no benchmarks available - data sets obtained during user tests were used. The visual representation offered by the fuzzy chart helps to reduce the cognitive complexity of the devices used in algorithmic music. The results obtained show that this approach is promising for future developments., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Comparison between composite index solution surfaces with fuzzy composite index decision surfaces

  • González Cárdenas, Rubén
  • Nebot Castells, M. Àngela
  • Múgica Álvarez, Francisco
Composite indices are used in many of the traditional approaches to measure risk to natural hazards. However, such indices are often built assuming linear interdependencies between the aggregated components, comprising in this way any realistic representation of the intricate and unseen processes that are behind each component along with their mutual influences when it comes to encapsulate a complex reality in a single construct. In this paper we used a Fuzzy Inference Systems type Mamdami to aggregate physical seismic risk and social vulnerability indicators without assuming linearity. The aggregation is made by establishing rules (if-then type) over the indicators in order to get a fuzzy composite index. We generated solution surfaces representing the outcome of a widely known linear aggregation method and the fuzzy system. Finally we performed a qualitative comparison between both to highlight their differences and possible consequences in terms of risk management., Peer Reviewed, Postprint (author's final draft)