comparison cross_sample/runCrossSample.xml @ 0:8d951baf795f draft

Uploaded

0:8d951baf795f

<tool id="run_cross_sample" name="Run Cross Sample" version="1.1">
  <description>using a Flow file that was run using FLOCK</description>
  <requirements>
    <requirement type="package" version="0.16.0">scipy</requirement>
    <requirement type="package" version="0.17.1">pandas</requirement>
    <requirement type="package" version="1.0">flock</requirement>
  </requirements>
  <stdio>
    <exit_code range="2" level="fatal" description="Columns inconsistencies between files and centroids file. See stderr for more details." />
  </stdio>
  <command><![CDATA[
    python $__tool_directory__/runCrossSample.py -m "${centroid}" -s "${stats}" -S "${mfistats}" -a "${allstats}" -M "${mfi}" -t $__tool_directory__ -o crossSampleOutputs
 #for $f in $input#
    -i $f
    -n "${f.name}"
 #end for#
 ]]>
  </command>
  <inputs>
    <param format="flowtext" name="input" type="data_collection" collection_type="list" label="Flowtext files Collection"/>
    <param format="flowmfi" name="centroid" type="data" label="Centroid file"/>
    <param name="mfi" type="select" label="Calculate centroids using:">
      <option value="mfi" selected="true">Mean Fluorescence Intensity</option>
      <option value="mdfi">Median Fluorescence Intensity</option>
      <option value="gmfi">Geometric Mean Fluorescence Intensity</option>
    </param>
  </inputs>
  <outputs>
    <collection type="list" label="CrossSample on ${input.name}" name="output">
      <discover_datasets pattern="(?P&lt;name&gt;.*)" directory="crossSampleOutputs" format="flowclr" />
    </collection>
    <data format="flowstat1" name="stats" label="Population distribution after CrossSample on ${input.name} using ${mfi}"/>
    <data format="flowstat2" name="mfistats" label="${mfi} centroids of CrossSample on ${input.name} using ${mfi}"/>
    <data format="flowstat3" name="allstats" label="${mfi} descriptive stats of CrossSample on ${input.name} using ${mfi}"/>
  </outputs>
  <tests>
    <test>
      <param name="input">
        <collection type="list">
          <element name="input1" value="input1.flowtext"/>
          <element name="input2" value="input2.flowtext"/>
          <element name="input3" value="input3.flowtext"/>
        </collection>
      </param>
      <param name="centroid" value="mfi.flowmfi"/>
      <param name="mfi" value="mfi"/>
      <output name="stats" file="out1.flowstat1" lines_diff="6"/>
      <output name="allstats" file="out1.flowstat3"/>
      <output name="mfistats" file="out1.flowstat2" compare="sim_size"/>
      <output_collection name="output" type="list" count="3">
        <element name="input1" file="run1/input1.flowtext.flowclr"/>
        <element name="input2" file="run1/input2.flowtext.flowclr"/>
        <element name="input3" file="run1/input3.flowtext.flowclr"/>
      </output_collection>
    </test>
    <test>
      <param name="input">
        <collection type="list">
          <element name="input1" value="input1.flowtext"/>
          <element name="input2" value="input2.flowtext"/>
          <element name="input3" value="input3.flowtext"/>
        </collection>
      </param>
      <param name="centroid" value="gmfi.flowmfi"/>
      <param name="mfi" value="gmfi"/>
      <output name="stats" file="out2.flowstat1" lines_diff="6"/>
      <output name="allstats" file="out2.flowstat3"/>
      <output name="mfistats" file="out2.flowstat2" compare="sim_size"/>
      <output_collection name="output" type="list" count="3">
        <element name="input1" file="run2/input1.flowtext.flowclr"/>
        <element name="input2" file="run2/input2.flowtext.flowclr"/>
        <element name="input3" file="run2/input3.flowtext.flowclr"/>
      </output_collection>
    </test>
  </tests>
  <help><![CDATA[
   This tool runs CrossSample using the MFI from FLOCK and text-converted FCS files.

-----

**Input**

This tool compares text-converted FCS files from a data collection to the MFI generated by a FLOCK run. The same data collection merged and run with FLOCK should be used to ensure consistency in the attribution of events to populations.

.. class:: infomark

The option chosen for the centroids (mean, median or geometric mean) should be the same as used to run FLOCK.

**Output**

Each event within each file of a dataset collection is attributed to a population depending on its intensity profile.
A table of the population composition of each file is generated as well as MFI and population descriptive statistics.

.. class:: infomark

Tip: If headers in each text-converted FCS file do not match those in the centroid file, the program will not run. Edit the input file using the Remove, rearrange and/or rename columns tool in the Flow File Tools section prior to Cross Sample analysis.

-----

**Example**

*Input* - fluorescence intensities per marker per event::

   Marker1 Marker2 Marker3
   33      47      11
   31      64      11
   21      62      99
   14      34      60

*Centroid file* - mean, geometric mean or median fluorescence intensity per marker per population::

   Population Marker1 Marker2 Marker3
   1          38      49      10
   2          21      63      100
   3          31      52      45
   4          11      78      25

*Output* for each text file - fluorescence intensities per marker and population ID per event::

   Marker1 Marker2 Marker3 Population
   33      47      11      1
   31      64      11      6
   21      62      99      2
   14      34      60      7

*Summary table* - counts of events in each population in each file::

   Filename SampleName Pop1 Pop2 Pop3 ...
   File1    Biosample1 255  147  18   ...
   File2    Biosample2 27   61   965  ...
   File3    Biosample3 201  546  221  ...
   File4    Biosample4 11   77   327  ...

*Centroid MFI Summary table* - for each file, mean, median or geometric mean fluorescence intensities per marker per population::

   Marker1 Marker2 Marker3 ... Population Percentage SampleName
   154     885     24      ... 1          0.2        Biosample1
   458     74      574     ... 2          0.3        Biosample1
   3       210     86      ... 3          0.05       Biosample1
   140     921     19      ... 1          0.1        Biosample2
   428     79      508     ... 2          0.25       Biosample2
   9       225     90      ... 3          0.3        Biosample2

*MFI Descriptive Statistics table* - for the set of files, mean, median and standard deviation of each centroid per marker per population, as well as mean, median and standard deviation of the population's proportion::

   Population Marker1_mean Marker1_median Marker1_stdev ... Percentage_mean Percentage_median Percentage_stdev
   1          94.65        90.86          25.8          ... 1.84            0.55              2.48
   2          132.18       131.58         5.02          ... 9.89            9.76              0.33
   3          71.8         69.68          10.53         ... 3.02            1.49              3.45
   4          84.85        84.85          nan           ... 8.52            8.52              nan
   5          161.82       132.77         61.29         ... 0.95            0.37              1.06
 ]]>
  </help>
  <citations>
    <citation type="doi">10.1002/cyto.b.20554</citation>
    <citation type="doi">10.3389/fimmu.2012.00302</citation>
    <citation type="doi">10.1371/journal.pone.0038408</citation>
    <citation type="doi">10.1371/journal.ppat.1003076</citation>
    <citation type="doi">10.1016/j.clim.2012.12.003</citation>
    <citation type="doi">10.1038/srep02327</citation>
  </citations>
</tool>