annotate cherri_train.xml @ 3:a30b9df1775a draft default tip

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/cherri commit 8778c3f71e295ec5b62ef3998319d211fc864e35

<tool id="cherri_train" name="Train a CheRRI model using RRIs" version="@VERSION@+galaxy@SUFFIX_VERSION@" profile="@PROFILE@">
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements"/>
    <command detect_errors="exit_code"><![CDATA[
        export PYTHONHASHSEED=31337 &&
        #set experiments = []
        #for $experiment in $rep_experiment:
            mkdir $experiment.exp_name &&
            mkdir '$experiment.exp_name/tmp' &&
            ln -s '$experiment.ref_source.genome_fasta' '$experiment.exp_name/genome.fa' &&
            $experiments.append(str($experiment.exp_name))
            #set replicates = []
            #for $i, $sample in enumerate($experiment.rep_samples):
                ln -s '$sample.file' '$experiment.exp_name/${i}.tabular' &&
                $replicates.append(str($i) + ".tabular")
            #end for
            cherri train
            -i1 '$experiment.exp_name'
            -r #echo ' '.join($replicates)
            -g '$experiment.exp_name/genome.fa'
            -l '$experiment.chrom_len_file'
            -n '$experiment.exp_name'
            #if $experiment.occupied_regions:
                -i2 '$experiment.occupied_regions'
            #end if
            -o .
            -on '$experiment.exp_name'
            -tp '$experiment.exp_name/tmp'
            @COMMONPARAMS@ &&
            #if len($rep_experiment) > 1:
                mkdir -p mixed_model &&
                ln -s '../$experiment.exp_name' 'mixed_model/$experiment.exp_name' &&
            #else:
                ln -s $experiment.exp_name/model/optimized/full_${experiment.exp_name}_context_${context}.model final_full.model &&
                #if $use_structure == 'off':
                    ln -s $experiment.exp_name/model/features/${experiment.exp_name}_context_${context}.npz features.npz &&
                #else:
                    ln -s $experiment.exp_name/feature_files/training_data_${experiment.exp_name}_context_${context}.npz features.npz &&
                #end if
            #end if
        #end for
        #if len($rep_experiment) > 1:
            mkdir mixed_model/tmp &&
            cherri train 
                -mi on
                -i1 mixed_model
                -r #echo ' '.join($experiments)
                -g /not/needed/
                -l /not/needed/
                -n mixed
                -o .
                -on mixed_model
                -tp mixed_model/tmp
                @COMMONPARAMS@ &&
                ln -s mixed_model/mixed/model/optimized/full_mixed_context_${context}.model final_full.model &&
                #if $use_structure == 'off':
                    ln -s mixed_model/mixed/model/features/mixed_context_${context}.npz features.npz &&
                #else:
                    ln -s mixed_model/mixed/feature_files/training_data_mixed_context_${context}.npz features.npz &&
                #end if
        #end if
        tar -zhcvf model.tgz final_full.model features.npz
    ]]></command>
    <inputs>
        <repeat name="rep_experiment" title="Experiment" min="1" default="1">
            <param name="exp_name" type="text" value="myExperiment" label="Name of the experiment" help="Only letters, numbers and underscores will be retained in this field. If more than one experiment is provided, then the tool generates a mixed model by combining datasets.">
                <sanitizer>
                    <valid initial="string.letters,string.digits"><add value="_" /></valid>
                </sanitizer>
            </param>
            <expand macro="reference_source_conditional"/>
            <repeat name="rep_samples" title="Interaction summary file" min="1" default="1">
                <param name="file" type="data" format="tabular" multiple="false" label="Interaction summary file of a replicate" />
            </repeat>
            <param name="occupied_regions" optional="True" type="data" format="bed" label="Path to the genomic RBP crosslink or binding site locations (in BED format)" />
        </repeat>
        <param name="context" type="integer" value="150" label="How much context should be added at up- and downstream of each sequence" />
        <param name="intarna_param_file" optional="True" type="data" format="txt" label="IntaRNA parameters file" />
        <param name="use_structure" type="boolean" truevalue="on" falsevalue="off" checked="true" label="Set 'off' if you want to disable structure, default 'on'" />
        <param name="run_time" type="integer" value="43200" label="Time used for the optimization in seconds, default: 43200 (12h)" />
        <param name="filter_hybrid" type="boolean" truevalue="on" falsevalue="off" checked="false" label="Filter the data for hybrids already detected by ChiRA" />
    </inputs>
    <outputs>
        <data name="out_model" format="tgz" from_work_dir="model.tgz" label="Trained model and features file on ${on_string}"/>
    </outputs>
    <tests>
        <!-- Single experiment -st on -->
        <test>
            <repeat name="rep_experiment">
                <param name="exp_name" value="myExperiment1"/>
                <conditional name="ref_source">
                    <param name="ref_source_selector" value="history"/>
                    <param name="genome_fasta" value="train_1.fa"/>
                </conditional>
                <param name="chrom_len_file" value="train_2_len.tabular" />
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
            </repeat>
            <param name="run_time" value="60" />
            <output name="out_model">
                <assert_contents>
                    <has_size min="100000"/>
                </assert_contents>
            </output>
        </test>
        <!-- Single experiment -st off -->
        <test>
            <repeat name="rep_experiment">
                <param name="exp_name" value="myExperiment1"/>
                <conditional name="ref_source">
                    <param name="ref_source_selector" value="history"/>
                    <param name="genome_fasta" value="train_1.fa"/>
                </conditional>
                <param name="chrom_len_file" value="train_2_len.tabular" />
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
            </repeat>
            <param name="run_time" value="60" />
            <param name="use_structure" value="off" />
            <output name="out_model">
                <assert_contents>
                    <has_size min="100000"/>
                </assert_contents>
            </output>
        </test>
        <!-- Mixed model -st on -->
        <test>
            <repeat name="rep_experiment">
                <param name="exp_name" value="myExperiment1"/>
                <conditional name="ref_source">
                    <param name="ref_source_selector" value="history"/>
                    <param name="genome_fasta" value="train_1.fa"/>
                </conditional>
                <param name="chrom_len_file" value="train_2_len.tabular" />
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
            </repeat>
            <repeat name="rep_experiment">
                <param name="exp_name" value="myExperiment2"/>
                <conditional name="ref_source">
                    <param name="ref_source_selector" value="history"/>
                    <param name="genome_fasta" value="train_2.fa"/>
                </conditional>
                <param name="chrom_len_file" value="train_2_len.tabular" />
                <repeat name="rep_samples">
                    <param name="file" value="train_2_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_2_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_2_pos.tabular" />
                </repeat>
            </repeat>
            <param name="run_time" value="60" />
            <param name="context" value="100" />
            <output name="out_model">
                <assert_contents>
                    <has_size min="100000"/>
                </assert_contents>
            </output>
        </test>
        <!-- Mixed model -st off -->
        <test>
            <repeat name="rep_experiment">
                <param name="exp_name" value="myExperiment1"/>
                <conditional name="ref_source">
                    <param name="ref_source_selector" value="history"/>
                    <param name="genome_fasta" value="train_1.fa"/>
                </conditional>
                <param name="chrom_len_file" value="train_2_len.tabular" />
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_1_pos.tabular" />
                </repeat>
            </repeat>
            <repeat name="rep_experiment">
                <param name="exp_name" value="myExperiment2"/>
                <conditional name="ref_source">
                    <param name="ref_source_selector" value="history"/>
                    <param name="genome_fasta" value="train_2.fa"/>
                </conditional>
                <param name="chrom_len_file" value="train_2_len.tabular" />
                <repeat name="rep_samples">
                    <param name="file" value="train_2_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_2_pos.tabular" />
                </repeat>
                <repeat name="rep_samples">
                    <param name="file" value="train_2_pos.tabular" />
                </repeat>
            </repeat>
            <param name="run_time" value="60" />
            <param name="use_structure" value="off" />
            <output name="out_model">
                <assert_contents>
                    <has_size min="100000"/>
                </assert_contents>
            </output>
        </test>
    </tests>
    <help><![CDATA[
        CheRRI in train mode to build a predictive model. Takes interactions summary files from ChiRA tool as input. Generates a model file and a features file that can be used in CheRRI eval mode.
    ]]></help>
    <expand macro="citations" />
</tool>