Inferring gene regression networks with model trees

Abstract

BackgroundNovel strategies are required in order to handle the huge amount of data produced by microarray technologies. To infer gene regulatory networks, the first step is to find direct regulatory relationships between genes building the so-called gene co-expression networks. They are typically generated using correlation statistics as pairwise similarity measures. Correlation-based methods are very useful in order to determine whether two genes have a strong global similarity but do not detect local similarities.ResultsWe propose model trees as a method to identify gene interaction networks. While correlation-based methods analyze each pair of genes, in our approach we generate a single regression tree for each gene from the remaining genes. Finally, a graph from all the relationships among output and input genes is built taking into account whether the pair of genes is statistically significant. For this reason we apply a statistical procedure to control the false discovery rate. The performance of our approach, named REGNET, is experimentally tested on two well-known data sets: Saccharomyces Cerevisiae and E.coli data set. First, the biological coherence of the results are tested. Second the E.coli transcriptional network (in the Regulon database) is used as control to compare the results to that of a correlation-based method. This experiment shows that REGNET performs more accurately at detecting true gene associations than the Pearson and Spearman zeroth and first-order correlation-based methods.ConclusionsREGNET generates gene association networks from gene expression data, and differs from correlation-based methods in that the relationship between one gene and others is calculated simultaneously. Model trees are very useful techniques to estimate the numerical values for the target genes by linear regression functions. They are very often more precise than linear regression models because they can add just different linear regressions to separate areas of the search space favoring to infer localized similarities over a more global similarity. Furthermore, experimental results show the good performance of REGNET.

Discovering gene association networks by multi-objective evolutionary quantitative association rules

“…We compared average values obtaining from modifying several input parameters. Finally, we set the input parameters and we present in Section 4.2.3 the performance of GarNet versus other benchmark methods reported in the literature [3,[9][10][11].…”

“…Consequently, to compare the inference capability of our approach against other methods we selected as benchmark methods: a decision-tree-based method [9], a regression-tree-based method [3] called RegNet, a probabilistic graphical model [10] and combinatorial optimization algorithms named GRNCOP [11] and GRN-COP2 [12].…”

“…Specifically, a decision-tree-based method [9], a regression-tree-based method [3], a probabilistic graphical model [10] and combinatorial optimization algorithm [12] are contrasted with the best ones inferred by GarNet.…”

“…This is a relevant constraint due to there might exist a strong local similarity over a subset of conditions which could not be detected with correlation measure. This constraint is taken into account in the model tree-based method proposed by Nepomuceno-Chamorro et al [3], where they used regression trees as a way to detect linear dependencies localized over a subset of conditions. Similar to this approach, another rule-based method is presented in [9] in which the authors used decision trees as a way to extract dependencies.…”

“…In general, these methods based on pairwise similarity measures are very useful to determine whether two genes have a strong global similarity under all conditions in the dataset. However, there could be strong local similarities over a subset of conditions, which could not be detected by them [3]. In this context, the discovery of Association Rules (AR), and particularly of Quantitative Association Rules (QAR), is a popular methodology that allows the discovery of significant and apparently hidden relations among attributes in a subspace of the instances from the dataset.…”