Titulo Estágio
PPRINT: Prediction of Protein-Protein Interactions
Opção Temática
Sistemas Inteligentes
Área Tecnológica
Informática Médica
Local do Estágio
DEI
Enquadramento
“If Edison had a needle to find in a haystack, he would examine straw after straw. I was sorry (…) knowing that a little theory and calculation would have saved him 90% of his labor.” – Nikola Tesla
Understanding life at a molecular level is complex, very complex. However it encloses a myriad of endless opportunities for mankind. Due to the large scale of the problem the use of appropriate computational tools and methods is of critical importance.
Toward this end networks are one of the dominant conceptualizations. Indeed they can be used to model relations and processes in a wide range of domains including computing, social, and natural sciences. A network is commonly defined as a collection of objects (nodes) connected by bidirectional links (edges) that usually have an associated weight and semantic.
Biomedical networks represent interactions among diverse biomolecular entities such as genes, proteins, glycans, or drugs. Several studies already endorse that the computational analysis of networks give insights about the role of genes in the cell. For instance the ‘central’ proteins, which topologically connect many different neighbourhoods of the network, are likely to mediate crucial biological functions. Given the interactomes evolved into this topology, analysing topological properties of biological networks should provide system-level insights on key players of biological processes.
Pavlopoulos et al, Using graph theory to analyze biological networks, BioData Mining, 2011, http://dx.doi.org/10.1186/1756-0381-4-10
Arrais et al, Using biomedical networks to prioritize gene–disease associations, Open Access Bioinformatics, 2010, http://dx.doi.org/10.2147/OAB.S21325
Coelho et al, From Protein-Protein Interactions to Rational Drug Design: Are Computational Methods Up to the Challenge?, special issue in Artificial Intelligence Techniques in Medicinal Chemistry, 2013
Objetivo
The current M.Sc. proposal aims at applying network principles and concepts to the biomedical domain being particularly focused in modelling disease-associated sub-networks in order to identify candidate proteins with high levels of association. This work program will require skills on information retrieval, semantic web, and data mining.
The expected outcome of this project will be a bioinformatics tool that presents a semantic graph and a term association methodology that address the issue of selecting the best candidates for gene-disease associations. The main challenges identified in the current state of the art and those wants to address in this project are:
• Lack of a centralized point for biomedical data and sparse associations between biomedical terms including semantic description;
• Low density of the biomedical network, including explicit relations;
• Tuning of graph algorithms for biomedical networks;
• Difficulty to fully explore the potentialities of semantics and association re-weighting.
Plano de Trabalhos - Semestre 1
1. Literature review on methods for network analysis
2. Study and test of tools for network representation
3. Development of a first working prototype
4. Initial evaluation of the developed method for detecting gene-disease association
5. Initial thesis writing
Plano de Trabalhos - Semestre 2
6. Final implementation and tuning of the method for detecting gene-disease association
7. Development of the bioinformatics tool for the analysis of gene-disease associations
8. Benchmark and comparative evaluation
9. Final thesis writing
Condições
There is currently no allowance available for this project. However the student will have access to the rooms of Adaptive Computation research group and to a cluster of computers. This is also an excellent opportunity to gain skills on Bioinformatics.
The students interested in this proposal are encouraged to contact the supervisors.
Observações
The student will benefit from a grant of 3000 Euros, paid in four monthly installments of 750 euros. This grant will be supported by CISUC.
He/she will also have access to the rooms of Adaptive Computation research group and to a cluster of computers. He/she will also have the opportunity to work in a multidisciplinary team. This is also an excellent opportunity to gain skills on Bioinformatics, an emergent field of research.
The students interested in this proposal are encouraged to contact the supervisor.
Orientador
Joel P. Arrais
jpa@dei.uc.pt 📩