miRNA gene regulatory networks for 38 human tissues (Q6688719)

We reconstructed miRNA regulatory networks for 38 tissues from the Genotype-Tissue Expression project (GTEx) using two different prior networks, one obtained with target predictions from TargetScan and one with target predictions from miRanda. We used these networks to investigate gene expression and regulation by miRNAs across these tissues. In the RData file, we share the following objects: - exp: a 16,161 by 9,435 data frame including normalized expression data for each sample. - expTS: a 16,161 by 38 matrix including the tissue-specificity scores for each gene in each tissue. - netT: a 10,391,523 by 41 data frame that includes the miRNA regulatory networks. The column miRNA includes the name of the regulating miRNA, the column Gene includes the target gene (HGNC symbol), and the column Prior the prior regulatory network based on target predictions from TargetScan, with 1 for edges that are canonical and 0 for edges that are non-canonical. The remaining 38 columns contain the PUMA network edge weights for each of the 38 tissues. - netT_TS: a 10,391,523 by 38 matrix that includes the tissue-specificity scores of the miRNA regulatory networks that were modeled on the TargetScan prior. Edges are not labelled, but edge order corresponds to the edges in netT. - netM: a 10,391,523 by 41 data frame that includes the miRNA regulatory networks. The column miRNA includes the name of the regulating miRNA, the column Gene includes the target gene (HGNC symbol), and the column Prior the prior regulatory network based on target predictions from miRanda, with 1 for edges that are canonical and 0 for edges that are non-canonical. The remaining 38 columns contain the PUMA network edge weights for each of the 38 tissues. - netM_TS: a 10,391,523 by 38 matrix that includes the tissue-specificity scores of the miRNA regulatory networks that were modeled on the miRanda prior. Edges are not labelled, but edge order corresponds to the edges in netT. - samples: a 9,435 by 2 data frame that includes sample identifiers (matching the identifiers in exp) and the tissue to which these samples belong. - mirnames: a 694 by 3 data frame that contains miRNA names of regulators and their matching target miRNA names. The first column base_miRNA contains the base miRNA, the name of the miRNA without any extensions. The second column reg_miRNA contains the 643 regulator miRNA, which may have -3P/-5P extensions, and which matches the miRNAs that are present as regulators in the networks. The third columns tar_miRNA contains the 621 target miRNAs, which may have numbered suffix extensions, and for which we have expression data available.