annotate glob_report.xml @ 3:79b9117aef01 draft

planemo upload for repository https://github.com/eteriSokhoyan/galaxytools/tree/branchForIterations/tools/GraphClust/CollectResults commit c03cf64554289eb098267c0923cf0cf7b245cc0c

<tool id="glob_report" name="Report_Results" version="0.1">
	<requirements>
		<requirement type="package" version="0.1">graphclust-wrappers</requirement>
		<requirement type="package" version='0.5'>perl-array-utils</requirement>
		<requirement type="package" version='0.18.1'>scikit-learn</requirement>
	</requirements>
	<stdio>
		<exit_code range="1:" />
	</stdio>
	<command>
		<![CDATA[
        unzip $FASTA  &> /dev/null &&

        #set $inputFiles = ""

        #for $cms_res in $cmsearch_results:
            #set $inputFiles += str($cms_res)+','
        #end for
        #set $inputFiles = $inputFiles[:-1]

        #set $inputFilesTrees = ""

        #for $mods in $model_tree_files:
            #set $inputFilesTrees += str($mods)+','
        #end for
        #set $inputFilesTrees = $inputFilesTrees[:-1]
		    glob_res.pl 
                '$inputFiles'
                $merge_cluster_ol
                $merge_overlap
                $min_cluster_size
                $cm_min_bitscore
                $cm_max_eval
                $cm_bitscore_sig
                $partition_type ''
                $cut_type
                '$inputFilesTrees'
        #if  $iteration_num.iteration_num_selector:
          $iteration_num.CI
          $final_partition_soft
          $final_partition_used_cmsearch
        #end if

        &&
        python '$__tool_directory__/evaluation.py'
		#if $cdhit:
    		&&
	    	python '$__tool_directory__/addCdhitseqs.py' '$cdhit'
	    #end if		
]]>
	</command>
	<inputs>
		<param type="data" name="FASTA" format="zip" />
		<param type="data" name="cmsearch_results" format="tabular" multiple="True"/>
		<param type="data" name="model_tree_files" format="txt" multiple="True"/>
		<param name="partition_type" type="boolean" checked="True" truevalue="0" falsevalue="1" label="Hard partition"/>
		<param name="cut_type" type="boolean" checked="True" truevalue="0" falsevalue="1" label="Use CM score for cutoff" help="otherwise use E-value"/>
		<param type="data" name="cdhit" format="txt" optional="true"/>
		<conditional name="iteration_num">
			<param name="iteration_num_selector" type="boolean"  checked="no" label="Multiple iterations"  help="for single iteration- NO, for multiple-YES"/>
			<when value="true">
				<param name="CI" type="integer" value="2" size="5" label="Number of current iteration "/>
				<param type="data" name="final_partition_soft" format="txt" />
				<param type="data" name="final_partition_used_cmsearch" format="txt" />
			</when>
			<when value="false" ></when>
		</conditional>
		<param name="merge_cluster_ol" type="float" value="0.66" size="5" label="merge_cluster_ol" help=""/>
		<param name="merge_overlap" type="float" value="0.51" size="5" label="merge_overlap" help=""/>
		<param name="min_cluster_size" type="integer" value="3" size="5" label="min_cluster_size" help=""/>
		<param name="cm_min_bitscore" type="integer" value="20" size="5" label="cm_min_bitscore" help=""/>
		<param name="cm_max_eval" type="float" value="0.001" size="5" label="cm_max_eval" help=""/>
		<param name="cm_bitscore_sig" type="integer" value="1" size="5" label="cm_bitscore_sig" help=""/>
	</inputs>
	<outputs>
		<data name="final_stats" format="txt" from_work_dir="RESULTS/cluster.final.stats" label="cluster.final.stats"  />
		<data name="tableForEval" format="tabular" from_work_dir="RESULTS/fullTab.tabular" label="tableForEval"  />
		<data name="final_soft" format="txt" from_work_dir="RESULTS/partitions/final_partition.soft" label="soft_part"   />
		<data name="final_used_cmsearch" format="txt" from_work_dir="RESULTS/partitions/final_partition.used_cmsearch" label="final_partition_used_cmsearch"   />
		<data name="evaluation" format="txt" from_work_dir="RESULTS/evaluation.txt" label="evaluation_of_clusters"   />
		<collection name="clusters" type="list" label="CLUSTERS">
			<discover_datasets pattern="(?P&lt;name&gt;^.*\.all$)" directory="RESULTS"  />
		</collection>
		<collection name="partitions" type="list" label="Partitions">
			<discover_datasets pattern="(?P&lt;name&gt;^.*$)" directory="RESULTS/partitions" />
		</collection>
	</outputs>
	<tests>
		<test>
			<param name="FASTA" value="FASTA.zip" ftype="searchgui_archive"/>
			<param name="cmsearch_results" value="1.tabular,2.tabular"/>
			<param name="model_tree_files" value="1.1.model.tree.fa,1.2.model.tree.fa"/>
			<param name="partition_type" value="0"/>
			<param name="cut_type" value="0"/>
			<conditional name="iteration_num">
				<param name="iteration_num_selector" value="false"/>
			</conditional>
			<param name="merge_cluster_ol" value="0.66"/>
			<param name="merge_overlap" value="0.51"/>
			<param name="min_cluster_size" value="3"/>
			<param name="cm_min_bitscore" value="20"/>
			<param name="cm_max_eval" value="0.001"/>
			<param name="cm_bitscore_sig" value="1"/>
			<output name="final_stats" file="RESULTS/cluster.final.stats" />
			<output_collection name="clusters" type="list">
				<element name="1.cluster.all" file="RESULTS/1.cluster.all" compare="contains"/>
				<element name="2.cluster.all" file="RESULTS/2.cluster.all" compare="contains"/>
			</output_collection>
			<output_collection name="partitions">
				<element name="final_overlap.map" file="RESULTS/partitions/final_overlap.map" compare="contains">
					<assert_contents>
						<has_text text="1.1  1.1" />
						<has_text text="1.2  1.2" />
					</assert_contents>
				</element>
				<element name="final_overlap.matrix" file="RESULTS/partitions/final_overlap.matrix" compare="contains">
					<assert_contents>
						<has_text text="MODEL CLASS 0 0" />
						<has_text text="1.2" />
						<has_text text="1.1" />
					</assert_contents>
				</element>
				<element name="final_partition.hard.best" file="RESULTS/partitions/final_partition.hard.best" />
				<element name="final_partition.hard.merged" file="RESULTS/partitions/final_partition.hard.merged" />
				<element name="final_partition.soft" file="RESULTS/partitions/final_partition.soft" />
				<element name="final_partition.used_cmsearch" file="RESULTS/partitions/final_partition.used_cmsearch" compare="contains"/>
			</output_collection>
		</test>
	</tests>
	<help>
		<![CDATA[

**What it does**

Post-processing. Redundant clusters are merged and instances that belong to multiple clusters
are assigned unambiguously. For every pair of clusters, the relative overlap (i.e. the fraction of
instances that occur in both clusters) is computed and clusters are merged if the overlap exceeds 50%.
Cluster members are finally ranked by their CM bitscore.

    ]]>
	</help>
	<citations>
    <citation type="bibtex">@inproceedings{costa2010fast,
        title={Fast neighborhood subgraph pairwise distance kernel},
        author={Costa, Fabrizio and De Grave, Kurt},
        booktitle={Proceedings of the 26th International Conference on Machine Learning},
        pages={255--262},
        year={2010},
        organization={Omnipress}
      }
      </citation>
  </citations>
</tool>