view edgeR_Concatenate_Expression_Matrices.xml @ 50:228867d5283b draft

Uploaded

<?xml version="1.0" encoding="UTF-8"?>
<tool id="concatenate_expression_matrices" name="edgeR: Concatenate Expression Matrices">
	<description>Create a full expression matrix by selecting the desired columns from specific count tables</description>
	
	<command>
		#set $j = 0
		#set $paste = []
		
		#if $add_geneids.choice == "true":
			#set $filename = str($j)+".txt"
			#set paste = paste + [$filename]
			
			cut -f $add_geneids.column_geneids.value "$add_geneids.sample_geneids" > $filename ;
			#set $j += 1
		#end if
		
		#for $sample in $samples:
			##echo "$sample.column_index"
			##echo "$sample.column_index.value"
			
			#set $column_str = ",".join([str(x).strip() for x in $sample.column_index.value])
			#set $filename = str($j)+".txt"
			#set paste = paste + [$filename]
			
			cut -f "$column_str" "$sample.sample" > $filename ;
			#set $j += 1
		#end for
		
		#if $add_lengths.choice == "true":
			#set $filename = str($j)+".txt"
			#set paste = paste + [$filename]
			
			cut -f $add_lengths.column_lengths.value "$add_lengths.sample_lengths" > $filename ;
			#set $j += 1
		#end if
		
		#set $paste_str = " ".join([str(x).strip() for x in $paste])
		
		paste $paste_str > $expression_matrix
	</command>
	
	<inputs>
		<conditional name="add_geneids">
			<param name="choice" type="select" label="Add a gene-IDs column at the end of the file" help="Highly recommended to select!" >
				<option value="false">No</option>
				<option value="true" selected="true">Yes</option>
			</param>
			<when value="false" />
			<when value="true">
				<param name="sample_geneids" type="data" format="tabular" label="Select Read-count dataset that contains a column for GeneIDs" help="from featureCounts/DEXSeq-count/HTSeq-count, etc." />
				<param name="column_geneids" numerical="false" use_header_names="true" label="Select GeneID column" type="data_column" data_ref="sample_geneids" multiple="false">
					<validator type="no_options" message="Please select at least one column." />
				</param>
			</when>
		</conditional>
		
		<repeat name="samples" title="Expression Table" help="E.g. a earlier concatenated table, or a HT-seq, featureCounts or DESeq-count result.">
			<param name="sample" type="data" format="tabular" label="Read-count dataset that belongs to a pair" help="from featureCounts/DEXSeq-count/HTSeq-count, etc." />
			<!-- select columns  -->
			<param name="column_index" numerical="false" use_header_names="true" label="Select columns that are associated with this factor level" type="data_column" data_ref="sample" multiple="true" size="120" display="checkboxes">
				<validator type="no_options" message="Please select at least one column." />
			</param>
		</repeat>
		
		<conditional name="add_lengths">
			<param name="choice" type="select" label="Add a gene-lengths column at the end of the file" help="Optional, only usefull if RPKM/FPKM calculation is desired." >
				<option value="false">No</option>
				<option value="true">Yes</option>
			</param>
			<when value="false" />
			<when value="true">
				<param name="sample_lengths" type="data" format="tabular" label="Read-count dataset that belongs to a pair" help="from featureCounts/DEXSeq-count/HTSeq-count, etc." />
				<param name="column_lengths" numerical="false" use_header_names="true" label="Select columns that are associated with this factor level" type="data_column" data_ref="sample_lengths" multiple="false" size="120">
					<validator type="no_options" message="Please select at least one column." />
				</param>
			</when>
		</conditional>
		
		<param name="remove_comment_lines" type="boolean" label="Automatically remove 'comment' lines starting with a '#'" truevale="true" falsevalue="false" selected="true" help="Some tools (incl. featureCounts) include comment lines that are not neccesairy for downstream analyis. By enabling this function, these lines will be removed." />
		
	</inputs>
	
	<outputs>
		<data format="tabular" name="expression_matrix" label="Expression matrix" />
	</outputs>
	
	<tests>
		<test>
			<param name="add_geneids.choice" value="true" />
			<param name="choice" value="true" />
			<param name="add_geneids" value="true" />
			
			
			<!--
Add a gene-IDs column at the end of the file	true	
Select Read-count dataset that contains a column for GeneIDs	21: MCF7_featureCounts_E2.txt	
Select GeneID column	1	
Read-count dataset that belongs to a pair	14: MCF7_featureCounts_Control_1.txt	
Select columns that are associated with this factor level	2	
Read-count dataset that belongs to a pair	15: MCF7_featureCounts_Control_2.txt	
Select columns that are associated with this factor level	2	
Read-count dataset that belongs to a pair	16: MCF7_featureCounts_Control_3.txt	
Select columns that are associated with this factor level	2	
Read-count dataset that belongs to a pair	17: MCF7_featureCounts_Control_4.txt	
Select columns that are associated with this factor level	2	
Read-count dataset that belongs to a pair	18: MCF7_featureCounts_Control_5.txt	
Select columns that are associated with this factor level	2	
Read-count dataset that belongs to a pair	22: MCF7_featureCounts_Control_6.txt	
Select columns that are associated with this factor level	2	
Read-count dataset that belongs to a pair	20: MCF7_featureCounts_Control_7.txt	
Select columns that are associated with this factor level	2	
Read-count dataset that belongs to a pair	21: MCF7_featureCounts_E2.txt	
Select columns that are associated with this factor level	2 3 4 5 6 7 8	
Add a gene-lengths column at the end of the file	true	
Read-count dataset that belongs to a pair	26: MCF7_featureCounts_GeneLengths.txt	
Select columns that are associated with this factor level	2	
Automatically remove 'comment' lines starting with a '#'	False
			-->
			
			<output name="expression_matrix" file="MCF7_featureCounts_concatenated.txt" />
		</test>
	</tests>
	
	<help>
edgeR: Concatenate Expression Matrices

**Notes**

Make sure the tables have an identical number of columns compared to the number of headers.
If you export tables using R, make sure you set: col.names=NA. Otherwise column may be swapped during concatenation.

**References**

The test data is coming from:  doi: 10.1093/bioinformatics/btt688.
http://www.ncbi.nlm.nih.gov/pubmed/24319002

	</help>
</tool>