<tool id="snmf" name="sNMF" version="2.0.0">
	<description>Fast and efficient program for estimating individual ancestry coefficients</description>
	<requirements>
                <requirement type="binary">perl</requirement>
                <requirement type="package" version="1.6.924">perl-bioperl</requirement>
		<requirement type="package" version="1.2">snmf</requirement>
		<requirement type="package" version="1.90b4">plink</requirement>
	</requirements>
        <stdio>
                <!-- [HELP] If no exit code rule is defined, the tool will stop if anything is written to STDERR -->
                <exit_code range="1:" level="fatal" />
        </stdio>
	<command interpreter="bash">./snmf.sh $vcf $outputs $logs $best_k_output $best_k_logfile $kmin $kmax $best_k_groups $threshold_group $entropy
        </command>
	<inputs>
		<param format="vcf" name="vcf" type="data" label="VCF file" help="VCF file"/>
		<param type="text" name="kmin" label="K min" value="2"/>
		<param type="text" name="kmax" label="K max" value="5"/>
		<param type="text" name="threshold_group" label="Minimum admixture proportion percentage for group assignation" value="50"/>
	</inputs>
	<outputs>
		<data format="txt" name="best_k_output" label="Best K Output"/>
		<data format="txt" name="best_k_groups" label="Best K Groups"/>
		<data format="txt" name="best_k_logfile" label="Best K Logfile"/>
		<data format="txt" name="outputs" label="Structure by sNMF"/>
		<data format="tabular" name="entropy" label="Cross-entropy values"/>
		<data format="txt" name="logs" label="All Logs"/>
	</outputs>

	<tests>
		<test>
			<param name="vcf" value="snmf_input.vcf" />
			<param name="kmin" value="2" />
			<param name="kmax" value="5" />
			<param name="threshold_group" value="50" />
			<output name="best_k_output" file="snmf_output"  compare="sim_size"  >
			    <assert_contents>
			   	<has_n_columns n="4" />
                                <has_line_matching expression=".*\s\d\..*" />
			    </assert_contents>
                        </output>
			<output name="best_k_groups" file="snmf_groups" compare="sim_size" >
			    <assert_contents>
                                <has_line_matching expression=".*;.*" />
			    </assert_contents>
                        </output>
			<output name="outputs" file="snmf_outputs.Q" compare="sim_size" >
			    <assert_contents>
                                <has_line_matching expression=".*\s\d\..*" />
			    </assert_contents>
                       </output> 

		</test>
	</tests>

	<help><![CDATA[

.. class:: infomark

**sNMF** version 1.2

.. class:: infomark

**Galaxy integration** Provided by Southgreen & Dereeper Alexis (IRD) & Marcon Valentin (IFB & INRA)

.. class:: infomark

**Support** For any questions about Galaxy integration, please send an e-mail to alexis.dereeper@ird.fr

---------------------------------------------------

====
sNMF
====

-----------
Description
-----------

  | Fast and efficient program for estimating individual admixture coefficients based on sparse non-negative matrix factorization and population genetics. 
  | For further informations, please visit the sNMF website_.


.. _website: http://membres-timc.imag.fr/Olivier.Francois/snmf/index.htm

------------
Dependencies
------------
sNMF
        snmf_ 1.2, Conda version
PLINK
        plink_ 1.90b4, Conda version
Bioperl
        perl-bioperl_ 1.6.924, Conda version

.. _snmf: https://anaconda.org/bioconda/snmf
.. _plink: https://anaconda.org/bioconda/plink
.. _perl-bioperl: https://anaconda.org/bioconda/perl-bioperl

	]]></help>
<citations>
       <citation type="doi">10.1534/genetics.113.160572</citation>
</citations>
</tool>