view mergeGenomicFiles.xml @ 37:e81019e3ac99

Updated synapseGetDataset to look at the filename rather than the (no longer existant) content type field to determine if the data is in zip format

<tool id="mergeGenomicFiles" description="Merge two genomic datasets into a new dataset" name="Merge Genomic Datasets" version="0.0.1">
  <description>
    Given two genomic datasets, merge them to create a larger dataset with the row and column identifiers from both datasets. Output this larger dataset, along with a 2-column matrix indicating the source file of each sample
  </description>
  <command interpreter="python">
      mergeGenomicMatrixFiles.py $inputA $inputB $outputC $outputSourceMatrix
      #if $labelForDatasetA
          --aLabel "${labelForDatasetA}"
      #end if
      #if $labelForDatasetB
          --bLabel "${labelForDatasetB}"
      #end if
  </command>
  <inputs>
    <param name="inputA" format="tabular" type="data" label="Genomic Dataset A"/>
    <param type="text" name="labelForDatasetA"  label="Dataset A Label (optional)" optional="true"/>
    <param name="inputB" format="tabular" type="data" label="Genomic Dataset B"/> 
    <param type="text" name="labelForDatasetB"  label="Dataset B Label (optional)" optional="true"/>
 </inputs>
  <outputs>
    <data name="outputSourceMatrix" format="tabular" label="Genomic Data Sources"/>
    <data name="outputC" format="tabular" label="Merged Genomic Data"/>
  </outputs>
  <help>
    ***Merge Genomic Datasets***

    Given two genomic datasets, merge them to produce a third dataset that is the union of the first two.  The new dataset will contain all column labels from either dataset, and all row labels from either dataset.  If a row label appears in both datasets, the output dataset will contain, for that row, all values for the first set of columns, plus all values for the second set of columns.  If a row label appears in the first dataset only, the output dataset will contain the values for the columns of the first dataset, and blanks (indicating missing values) for the columns of the second dataset.

    To maintain provenance, this script also outputs a second matrix, with one row for each column in the output dataset, and two columns per row indicating which input dataset that column came from.  By default, the input dataset name is used to indicate which input file each column came from.  Optionally, the user can specify descriptive labels to be used in place of the filenames.  This all assumes that each column exists in only one input dataset.
  </help>
</tool>