<tool id="graphclust_aggregate_alignments" name="Aggregate and filter alignment metrics" version="0.1">
 <description>of individual clusters, like the output of graphclust_align_cluster</description> 
    <requirements>
        <requirement type="package" version="0.6.0">graphclust-wrappers</requirement>
        <requirement type="package" version="0.23.0">pandas</requirement>
    </requirements>
    <command detect_errors="exit_code">
    <![CDATA[
        mkdir ./clusters_metrics &&
        #import re    
        #for $tsv in $alignment_cons_tsv:
            #set $safename_tsv = re.sub('[^\w\-_\.]', '_', $tsv.element_identifier)
            ln -f -s  '$tsv' ./clusters_metrics/${safename_tsv}.tsv &&
        #end for
        aggregate_align_metrics.py 
        --RNAz-prob-threshold $RNAz_prob_threshold  
        --rscape-bp-threshold $rscape_bp_threshold 
        --min-seq-num $min_seq_num
        $exclude_spurious_structs
        $additonal_columns
        --clusters-tsv-pattern 
        "clusters_metrics/*.tsv" 
        --filtered-tsv-out filtered-alignment-metrics.tsv
        --bed-out bed-cluster-locations.bed
            
    ]]></command>
    <inputs>
        <param type="data" name="alignment_cons_tsv" format="tabular" multiple="True" label="conservation-metrics-tsv"
            help="Tabular tsv file of the computed conservation metrics from align cluster step"/>
        <param argument="min_seq_num" type="integer" value="3"   label="Minimum cluster size" 
            help="Clusters with fewer seqeunces are excluded"/>
        <param argument="RNAz_prob_threshold" type="float" value="0.5" label="RNAz bed threshold" 
            help="Minimum RNAz SVM RNA-class probablity for bed annotation as RNAz hit"/>
        <param argument="rscape_bp_threshold" type="integer" value="2" label="Rscape bed threshold"
            help="Minimum number of significant covarying basepairs by Rscape for bed annotation as Rscape hit"/>
        <param name="exclude_spurious_structs" type="boolean" checked="True" truevalue="--exclude-spurious-structs" falsevalue="" 
            help="Exclude spurious alignment structs with poor SCI less than 0.01"/>
        <param name="additonal_columns" type="boolean" checked="False" truevalue="--all-columns" falsevalue="" help="Output additional prediction metrics"/>
    </inputs>
    <outputs>
        <data name="filtered_tsv" format="tabular" from_work_dir="filtered-alignment-metrics.tsv" label="filtered-alignment-metrics.tsv"  />
        <data name="bed_clusters" format="tabular" from_work_dir="bed-cluster-locations.bed" label="bed-cluster-locations.bed"  />
    </outputs>
    <tests>
        <test>
            <param name="alignment_cons_tsv" value="1.alignment.cons.tsv,2.alignment.cons.tsv,3.alignment.cons.tsv"/>            
            <output name="filtered_tsv" file="metrics1.tsv"  />
            <output name="bed_clusters" file="clusters1.bed"  />
        </test>
    </tests>
    <help><![CDATA[
Aggregate and filter alignment metrics of individual clusters, like the output
of graphclust_align_cluster
    ]]></help>

  <citations>
    <citation type="doi">10.5281/zenodo.597695</citation>
  </citations>

</tool>