Advisor PhD: David Corrêa Martins Junior
Advisor MsC: Marcelo Zanchetta do Nascimento
Location: São Paulo, Brazil
Contact: ricardo.jacomini [at] usp.br
Research
Ricardo PhD project aims for modeling, analyzing and inferring gene networks with biological characteristics. It is important to observe that this approach is only part of a complex system, it can be used as a starting point to understanding the basis of some diseases and the significance of the genetic variation, by analyzing the genes complexity and behavior. It is noteworthy that currently there are only a small number of available sequenced databases, with benchmarks in order to validate new computational models. Important investigations in the literature concerning modeling and identification of gene networks use DNA micro-array data expression in order to quantitatively recognize patterns in data. However, analyzing a large amount of biological data is still a very difficult task.
Some researchers have applied statistical approaches, while others perform a molecular and functional analysis of encoded and regulated genes. Usually, given the complexity of such data, the relationships of genes are examined in networks. However, analyzing these networks can be a complex task, and the visualization of biological networks is one of the main problems in this domain. Especially, because a gene is encoded by one, two or more genes. Thus, one approach to reduce the complexity is to search for relationships of genes through the dynamics of the signs of gene expression data.
This project is being applied simulation of gene expression data and artificial intelligence techniques. An algorithm for multi-resolution analysis will be developed, aiming to model the dynamics of the signals involved in the regulatory process. Genes with similar biological functions will be grouped, defining different levels of similarity between predictors genes and targets genes. It is expected an algorithm more efficient for features selection to infer the predictors genes in relation to one target genes, combining a method of multi-variate and multi-resolution analysis. After that will be applied validation metrics to this methodology, using an In Silico benchmark of gene expression data. The results of this project will assist professionals in biomedical and related areas in decision-making in relation to the dynamics of the control of gene regulatory systems in real databases.
Keywords: Multivariate and Multiresolution Analysis, Probabilistic Gene Networks, Statistical Learning.
Biography
Ricardo de Souza Jacomini was born in Santo André Brazil, in 1974. He received the B.S. in Mathematics from Fundação Santo André and the M.Sc. degree from the Federal University of ABC, Santo André, Brazil, in 2002 and 2011, respectively. He is currently a PhD Student in Eletrical Engineering at the São Paulo University, São Paulo, Brazil. His main research interests are Machine Learning applied to Systems Biology, Pattern Recognition, Digital Signal Processing and Computer Vision.
Computer Skills
• Servers Linux , Java and QT / C++, Matlab, R Language.
• Information Security, Communication Networks.
• Frequent online learner (Coursera, Stanford, etc.)