Loading...
Loading...

Go to the content (press return)

Computational Intelligence for Knowledge Discovery from G Protein-Coupled Receptors

Total activity: 18
Type of activity
Competitive project
Acronym
COIN-GPCR
Funding entity
MIN DE ECONOMIA Y COMPETITIVIDAD
Funding entity code
TIN2016-79576-R
Amount
47.069,00 €
Start date
2016-12-30
End date
2020-12-29
Abstract
Current biology has become a strongly data-dependent science, mostly due to the rapid advances in the omics sciences. The overabundance
of data in the field makes the development of suitable Knowledge Discovery (KD) strategies a paramount goal. Computational
Intelligence (CI) and, in particular, Machine Learning (ML) have been identified as key analytical approaches to KD in the omics sciences.
The broad context of this proposal is the field of pharmacoproteomics: the application of proteomic science to drug development and
assessment. This is also a data-intensive area, with great potential for knowledge-based economies. According to the 2014 Cotec:
Technology and Innovation in Spain report, the Spanish pharmaceutical industry is a sector at the forefront of R+D.
The proposal builds on the results of our previous project "Knowledge Acquisition in Pharmacoproteomics using advanced Artificial
Intelligence" and focuses on a type of cell membrane proteins: the G Protein-Coupled Receptors (GPCR). Over 50% of current drugs
target just four families of proteins, of which almost 30% are GPCRs. Their Class C, object of analysis in the proposed project, is the target
of drugs for pain, anxiety and neurodegenerative disorders.
The functional properties of proteins depend on their tertiary structure, which informs of their 3-D configuration. Progress in the discovery
of the 3-D structure of GPCRs is slow and recent, specially for Class C: it was not until 2014 that the partial tertiary structures of two of
them were discovered. Class C GPCRs are thus usually studied from their primary amino acid sequences, which are publicly available
from international data repositories. Class C can be characterized at different levels of detail, but their "subtype labels" and the assignment
of given sequences to subtypes are far from being universally accepted.
In this proposal, we take on the challenge of analyzing Class C GPCR primary sequences using CI methods. These include
implementations of state-of-the-art Deep Learning (DL) approaches, which only over the last few years have started to be used in the area
of proteomics, and also of well-established techniques specifically tailored for the type of sequential data analyzed in the project.
This leaves several avenues open to research that we address from two different and transversal points of view: On the one hand, and
from a data science perspective, we aim to develop novel and improve existing CI-based techniques to tackle these problems, including
DL for the direct analysis of symbolic GPCR sequences; the definition of biologically-plausible kernels for discrete sequential data; and
statistical ML and fuzzy methods for supervised and unsupervised hierarchical subtyping of GPCRs.
On the other hand, and from a bioinformatics perspective, we aim to investigate problems of biological interest such as the search for
motifs and binding sites in the sequence; the analysis of the different roles of the extra-cellular/intra-cellular/transmembrane parts of the
receptor on subtype characterization at different levels of detail; label noise analysis for database quality assessment; and the analysis of
the effects of activation / inhibition of Class C receptors produced by small biomolecules.
In our endeavor, we count with the expert support and collaboration of the Laboratory of Molecular Neuropharmacology and Bioinformatics
at Universitat Autònoma de Barcelona.
Keywords
Aprendizaje Automático, Bioinformatics, Bioinformática, Computational Intelligence, Fuzzy Systems, Fármaco-proteómica, GPCR, Innovation; Health, Machine Learning, Pharmacoproteomics, Sistemas Borrosos, Wellness & Inclusion; Inteligencia Computacional
Scope
Adm. Estat
Plan
Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016
Resoluton year
2016
Funcding program
Programa Estatal de I+D+i Orientada a los Retos de la Sociedad
Funding call
Retos de Investigación: Proyectos de I+D+i
Grant institution
Gobierno De España. Ministerio De Economía Y Competitividad, Mineco

Participants

Scientific and technological production

1 to 18 of 18 results