Banca de DEFESA: RODRIGO CESAR BONINI

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
STUDENT : RODRIGO CESAR BONINI
DATE: 01/11/2024
TIME: 14:00
LOCAL: por participação remota em: https://conferenciaweb.rnp.br/sala/david-41
TITLE:

Gene Networks Inference by Deep Reinforcement Learning

PAGES: 77
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
SUMMARY:

Gene Regulatory Networks (GRN) inference from gene expression data is an important problem in systems biology field, involving the estimation of gene-gene indirect dependencies and the regulatory functions among these interactions to provide a model that explains the gene expression dataset. The main goal is to comprehend the global molecular mechanisms underlying diseases for the development of medical treatments and drugs. However, such a problem is considered an open problem, since it is difficult to obtain a satisfactory estimation of the dependencies given a very limited number of samples subject to experimental noises. Many gene networks inference methods have been proposed in the literature, where some of them use heuristics or model based algorithms to find interesting networks that explain the data by codifying whole networks as solutions. However, in general, these models are slow, not scalable to real sized networks (thousands of genes), or require many parameters, the knowledge from an specialist or a large number of samples to be feasible. On the other hand, Reinforcement Learning is an adaptable goal oriented approach that does not require large labeled datasets and many parameters; can give good quality solutions in a feasible execution time; and can work automatically for a long time without the need of a specialist, but still needs some improvements to deal with its scalability. Therefore, this thesis propose a way to adapt Reinforcement Learning to the Gene Regulatory Networks inference domain, making an improvement in its performance, using Deep Reinforcement Learning, a framework that approximate and can provide better solutions with less computational effort. Hence, we can get networks with quality comparable or even superior to ones achieved by exhaustive search, but in much smaller execution time. An experimental evaluation shows that the proposal of this thesis is promising in learning and successfully finding good solutions across different tasks automatically in a reasonable time, scaling well for real-sized networks containing hundreds or thousands of genes. Among the future perspectives opened by this work, the method can be improved with possible more efficient implementations with regard to memory consumption and the employment of parallel computing.

COMMITTEE MEMBERS:
Presidente - Interno ao Programa - 1722875 - DAVID CORREA MARTINS JUNIOR
Membro Titular - Examinador(a) Interno ao Programa - 2078059 - LUIZ ANTONIO CELIBERTO JUNIOR
Membro Titular - Examinador(a) Externo ao Programa - 1675615 - LUIZ CARLOS DA SILVA ROZANTE
Membro Titular - Examinador(a) Externo à Instituição - SERGIO NERY SIMÕES - IFES
Membro Titular - Examinador(a) Externo à Instituição - RICARDO DE SOUZA JACOMINI - Johns Hopkin
Membro Suplente - Examinador(a) Interno ao Programa - 1673092 - RONALDO CRISTIANO PRATI
Membro Suplente - Examinador(a) Externo à Instituição - THIAGO PEDRO DONADON HOMEM - IFSP
Membro Suplente - Examinador(a) Externo à Instituição - FELIPE LENO DA SILVA