Extreme learning machines for discovering gene regulatory networks from temporal profiles of expression

RUBIOLO, M.; MILONE, D.H.; STEGMAYER, G.

Conferencia; 4th ISCB-LA Bioinformatics Conference; 2016

Reconstructing a gene regulatory network (GRN) from gene expression data is one of theultimate goals of bioinformatics today. In recent years, several computational methods wereproposed to support the discovery of GRN, with different levels of accuracy. Most of themrequire several input sources to provide an acceptable prediction. Indeed, not only geneexpression but also experimental results such as the wild type unperturbed network, steadystate levels of singlegene knockouts and knockdowns are required. Thus, there is an important challenge in reconstructing a GRN only from temporal geneexpressiondata. We present here a model capable of modeling all possible genetogeneregulations just from gene expression data by using a pool of fast and accurate artificial neural networks.