Galaxy | Tool Preview

Aurora Galaxy WGCNA (version 1.0.0)
The gene expression data is an n x m matrix where n rows are the genes, m columns are the samples and the elements represent gene expression levels (derived either from Microarray or RNA-Seq). The matrix should be stored in a comma-separated (CSV) file and it must have a header. The gene names must appear as the first column of data in the file.
Within the gene expression data some genes may have missing values. If so, please indicate the text that is used to identify a missing value. Some common examples include: NA, 0.0, 0, -Inf. Any expression level that exactly matches the value provided will be considered a missing value.
When checking for outliers, WGCNA performs hierarchical clustering. The resulting dendrogram can be cut at the given height to remove outliers. If no value is provided a cut height will automatically be determined. Try running this tool first without providing a value. Return and set a value if the results are not adequate.
Prior to network construction, WGCNA recommends that the gene expression data is raised to a power. The exact power that should be used will be automatically determined. Try running this tool first without providing a value. Return and set a value if the results are not adequate.
The minimum module size. Modules smaller than this will not be included in the network.
Constructing a network can use an extreme amount of memory if the number of genes is high. The block size enables WGCNA to divide the data into blocks of genes with similar expression reducing the amount of memory used. The block size indicates the maximum number of genes that can be used in a block. The total number of blocks used will be the total genes divided by this number (plus 1 for any remainder).
While WGCNA uses a soft thresholding approach for finding modules and constructing gene similarity, when exporting the network for display as a graph a hard threshold is still required. For WGCNA, the threhshold is applied to the Euclidian distance between all genes. But, there is no set prescribed method to decide on a proper hard threshold value. Set a threshold now, then you can apply filters later (such as in Cytoscape) to remove low weighted edges if desired.
Trait/Phenotypes
Trait/Phenotype 0
This tool is a wrapper for the WGCNA R library. Please see the online WGCNA tutorial for further details.