Mercurial > repos > bgruening > deseq2
diff deseq2.xml @ 17:a05999fd6e26 draft
Uploaded
| author | bgruening |
|---|---|
| date | Mon, 30 Sep 2013 10:51:41 -0400 |
| parents | 1d2a02bc2208 |
| children | 72613128bd65 |
line wrap: on
line diff
--- a/deseq2.xml Mon Sep 09 07:09:38 2013 -0400 +++ b/deseq2.xml Mon Sep 30 10:51:41 2013 -0400 @@ -1,5 +1,5 @@ <tool id="deseq2" name="DESeq2" version="2.0.1"> - <description>Determines differentially expressed transcripts from read alignments</description> + <description>Determines differentially expressed features from count data</description> <requirements> <requirement type="binary">Rscript</requirement> <requirement type="R-module">DESeq2</requirement> @@ -7,33 +7,33 @@ <requirement type="set_environment">DESEQ2_SCRIPT_PATH</requirement> </requirements> <command interpreter="Rscript"> + #import simplejson deseq2.R -o "$deseq_out" + --outfilefiltered "$deseq_out_filtered" #if $pdf: - -p "$plots" + -p "$plots" #end if - + -i "$input_matrix" - #if $filter_sel.filter_sel_opts == 'all_vs_all': - -s 'all_vs_all' - #else: - -s ## build a string like '1,2 5,6' - "${filter_sel.control_cols} ${filter_sel.experiement_cols}" + #set $temp_factor_name = list() + #for $factor in $rep_factorName: + #set $temp_factor = dict() + #for $level in $factor.rep_factorLevel: + ##$temp_factor_list.append( '%s::%s:%s' % ($factor.factorName.replace(' ','_'), $level.factorLevel, $level.factorIndex) ) + $temp_factor.update({str($level.factorLevel): map(int, str($level.factorIndex).split(','))}) + #end for + $temp_factor_name.append([str($factor.factorName), $temp_factor]) - #set $temp_factor_list = list() - #set $is_multi_factor_analysis = False - #for $factor in $filter_sel.factor: - #set $is_multi_factor_analysis = True - $temp_factor_list.append( '%s:%s' % ($factor.factor_name.replace(' ','_'), $factor.factor_index) ) #end for - #if $is_multi_factor_analysis: - -f "#echo ' '.join( $temp_factor_list )#" - #end if - #end if - -t $fittype + ##-m "#echo ' '.join( $temp_factor_list )#" + -m '#echo simplejson.dumps(temp_factor_name)#' + ##--organism "$organism" + ##-t "$fittype" + -c $countthreshold </command> <stdio> <regex match="Execution halted" @@ -51,56 +51,64 @@ </stdio> <inputs> <param format="tabular" name="input_matrix" type="data" label="Countmatrix" help="You can create a count matrix with the tool - 'Count reads in features with htseq-count'"/> + 'Count reads in features with htseq-count'"/> + + <repeat name="rep_factorName" title="Factor/Condition" min="1"> + <param name="factorName" type="text" value="FactorName" label="Specify a factor name" help=""/> + <repeat name="rep_factorLevel" title="Factor level" min="1"> + <param name="factorLevel" type="text" value="FactorLevel" label="Specify a factor level" help=""/> - <conditional name="filter_sel"> - <param name="filter_sel_opts" type="select" label="Select mode of analysis" help="All against all: analyses all possible pairs of conditions, - <br> Specific: select your samples of interest"> - <option value="all_vs_all">All against all</option> - <option value="specific">specific</option> - </param> - <when value="specific"> - - <param name="control_cols" label="Select columns containing first condition" type="data_column" data_ref="input_matrix" - numerical="True" multiple="true" use_header_names="true" size="120" display="checkboxes"> - <validator type="no_options" message="Please select at least one column."/> - </param> - - <param name="experiement_cols" label="Select columns containing second condition" type="data_column" data_ref="input_matrix" + <param name="factorIndex" label="Select columns that are associated with this factor level" type="data_column" data_ref="input_matrix" numerical="True" multiple="true" use_header_names="true" size="120" display="checkboxes"> <validator type="no_options" message="Please select at least one column."/> </param> - - <repeat name="factor" title="Include factor"> - <param name="factor_name" type="text" value="Factor Name" label="Specify a factor name" help=""/> - - <param name="factor_index" label="Select columns that are associated with a factor." type="data_column" data_ref="input_matrix" - numerical="True" multiple="true" use_header_names="true" size="120" display="checkboxes"> - <validator type="no_options" message="Please select at least one column."/> - </param> - </repeat> - - </when> - <when value="all_vs_all" /> - </conditional> - + </repeat> + </repeat> + <!-- + <param name="control_cols" type="select" display="checkboxes" multiple="true" optional="True" label="Select columns containing first condition" + dynamic_options="get_matrix_header( input_dataset=input_matrix )" help="insert useful info here"> + <validator type="no_options" message="Please select at least one column."/> + </param> + <param name="experiement_cols" type="select" display="checkboxes" multiple="true" optional="True" label="Select columns containing second condition" + dynamic_options="get_matrix_header( input_dataset=input_matrix )" help="insert useful info here"> + <validator type="no_options" message="Please select at least one column."/> + </param> + + <repeat name="factor" title="factor"> + <param name="factor_name" type="text" value="Factor Name" label="Specify a factor name" + help="Order of factors you add effects the design formual and hence effects whole analysis. Design formual will be created as follows: ~factor1+factor2+factor3+...+factorN+condition"/> + <param name="factor_index" type="select" display="checkboxes" multiple="true" optional="True" label="Choose sample to compare with" + dynamic_options="get_matrix_header( input_dataset=input_matrix )" help="Select columns that are associated with a factor"> + <validator type="no_options" message="Please select at least one column."/> + </param> + </repeat> + --> + <!--param name="organism" size="10" type="select"> + <option value="human">human</option> + <option value="mouse">mouse</option> + <option value="fly">fly</option> + <option value="other">other</option> + </param--> + <param name="countthreshold" size="10" type="float" value="10" label="Filter out features with mean normalized counts lower than this value"/> <param name="fittype" type="select" label="Type of fitting of dispersions to the mean intensity"> - <option value="parametric">parametric</option> - <option value="local">local</option> - <option value="mean">mean</option> - </param> - <param name="pdf" type="boolean" truevalue="" falsevalue="" checked="true" label="Visualising the analysis results" - help="output an additional PDF file" /> + <option value="parametric">parametric</option> + <option value="local">local</option> + <option value="mean">mean</option> + </param> + <param name="pdf" type="boolean" truevalue="" falsevalue="" checked="true" label="Visualising the analysis results" + help="output an additional PDF files" /> </inputs> - <outputs> - <data format="tabular" name="deseq_out" label="DESeq2 result file on ${on_string}"/> - <data format="pdf" name="plots" label="DESeq2 plots on ${on_string}"> - <filter>pdf == True</filter> - </data> - </outputs> + <outputs> + <data format="tabular" name="deseq_out" label="DESeq2 result file on ${on_string}"/> + <data format="tabular" name="deseq_out_filtered" label="Independent filtering result file on ${on_string}"/> + <data format="pdf" name="plots" label="DESeq2 plots on ${on_string}"> + <filter>pdf == True</filter> + </data> + </outputs> + <code file="matrix_helper.py" /> - <help> + <help> .. class:: infomark @@ -122,13 +130,12 @@ ====== ========================================================== Column Description ------ ---------------------------------------------------------- - 1 Condition tested (corresponds to the conitions from the first line of your count matrix) - 2 Gene Identifiers - 3 mean normalised counts, averaged over all samples from both conditions - 4 the logarithm (to basis 2) of the fold change - 5 standard error estimate for the log2 fold change estimate - 6 p value for the statistical significance of this change - 7 p value adjusted for multiple testing with the Benjamini-Hochberg procedure + 1 Gene Identifiers + 2 mean normalised counts, averaged over all samples from both conditions + 3 the logarithm (to basis 2) of the fold change + 4 standard error estimate for the log2 fold change estimate + 5 p value for the statistical significance of this change + 6 p value adjusted for multiple testing with the Benjamini-Hochberg procedure which controls false discovery rate (FDR) ====== ========================================================== @@ -150,5 +157,5 @@ .. _DESeq2: http://master.bioconductor.org/packages/release/bioc/html/DESeq2.html - </help> + </help> </tool>