<tool id="custom_pro_db" name="CustomProDB" version="1.16.1.0">
  <description>Generate protein FASTAs from exosome or transcriptome data</description>
  <requirements>
    <requirement type="package" version="3.3.1">r-base</requirement>
    <!--<requirement type="package" version="1.14.0">bioconductor-customprodb</requirement>-->
    <requirement type="package" version="1.18.0">bioconductor-rgalaxy</requirement>
    <requirement type="package" version="1.21.0">bioconductor-biocinstaller</requirement>
    <requirement type="package" version="1.20.3">bioconductor-variantannotation</requirement>
    <requirement type="package" version="1.26.4">bioconductor-genomicfeatures</requirement>
    <requirement type="package" version="1.11.1">r-devtools</requirement>
    <requirement type="package" version="3.98_1.4">r-xml</requirement>
    <requirement type="package" version="0.10.11">r-rmysql</requirement>
    <requirement type="package" version="1.0.2">r-testthat</requirement>
    <requirement type="package" version="0.1.0">r-getoptlong</requirement>
    <requirement type="package" version="1.1.2">r-stringi</requirement>
    <requirement type="package" version="1.1.0">r-stringr</requirement>
    <requirement type="package" version="1.10.0">r-data.table</requirement>
    <requirement type="package" version="0.4_10">r-sqldf</requirement>
    <requirement type="package" version="0.6_6">r-gsubfn</requirement>
    <requirement type="package" version="2.3_47">r-chron</requirement>
    <requirement type="package" version="0.3_10">r-proto</requirement>
    <requirement type="package" version="1.8.4">r-plyr</requirement>
  </requirements>
  <stdio>
    <exit_code range="1:" level="fatal" description="Job Failed" />
  </stdio>
  <command><![CDATA[
    Rscript --vanilla '$__tool_directory__/customProDB.R'
       --bam='$genome_annotation.bamInput'
       --bai='${genome_annotation.bamInput.metadata.bam_index}'
       --vcf='$genome_annotation.vcfInput'
       --rpkmCutoff=$rpkmCutoff
       $outputIndels
       $outputSQLite
       $outputRData

       #if str($genome_annotation.source) == 'history':
            --exon_anno='$genome_annotation.exonAnno'
            --proteinseq='$genome_annotation.proteinSeq'
            --procodingseq='$genome_annotation.proCodingSeq'
            --ids='$genome_annotation.ids'
            #if str($genome_annotation.dbsnpInCoding) != 'None':
                --dbsnpinCoding='$genome_annotation.dbsnpInCoding'
            #end if
            #if str($genome_annotation.cosmic) != 'None':
                --cosmic='$genome_annotation.cosmic'
            #end if
       #else:
            #set index_path = $genome_annotation.builtin.fields.path
            --exon_anno='$index_path/exon_anno.RData'
            --proteinseq='$index_path/proseq.RData'
            --procodingseq='$index_path/procodingseq.RData'
            --ids='$index_path/ids.RData'
            #if $genome_annotation.dbsnpInCoding:
                --dbsnpinCoding='$index_path/dbsnpinCoding.RData'
            #end if
            #if $genome_annotation.cosmic:
                --cosmic='$index_path/cosmic.RData'
            #end if
       #end if
    2>&1
]]>
  </command>
  <inputs>
    <conditional name="genome_annotation">
      <param name="source" type="select" label="Will you select a genome annotation from your history or use a built-in annotation?" help="See `Annotations` section of help below">
        <option value="builtin">Use a built-in genome annotation</option>
        <option value="history">Use annotation from your history</option>
      </param>
      <when value="builtin">
        <param name="builtin" type="select" label="Select genome annotation" help="If your genome of interest is not listed, contact the Galaxy team">
          <options from_data_table="customProDB">
            <filter type="sort_by" column="2"/>
            <validator type="no_options" message="No annotations are available for the selected input dataset"/>
          </options>
        </param>
        <param name="bamInput" type="data" format="bam" label="BAM file">
            <validator type="unspecified_build" />
            <validator message="Sequences are not currently available for the specified build." metadata_column="1" metadata_name="dbkey" table_name="customProDB" type="dataset_metadata_in_data_table" />
        </param>
        <!--<param name="baiInput" type="data" format="bam_index" label="BAM Index (BAI) file">
            <validator type="unspecified_build" />
            <validator message="Sequences are not currently available for the specified build." metadata_column="1" metadata_name="dbkey" table_name="customProDB" type="dataset_metadata_in_data_table" />
        </param>-->
        <param name="vcfInput" type="data" format="vcf" label="VCF file">
            <validator type="unspecified_build" />
            <validator message="Sequences are not currently available for the specified build." metadata_column="1" metadata_name="dbkey" table_name="customProDB" type="dataset_metadata_in_data_table" />
        </param>
        <param name="dbsnpInCoding" type="boolean" value="" label="Annotate SNPs with rsid from dbSNP" help="The genome annotation must have been prepared with dbSNP." />
        <param name="cosmic" type="boolean" value="" label="Annotate somatic SNPs from COSMIC (human only)" help="The genome annotation must have been prepared with COSMIC." />
      </when>
      <when value="history">
        <param name="exonAnno" type="data" format="RData" help="A dataframe of exon annotations in an RData file" label="Exon Annotations" />
        <param name="proteinSeq" type="data" format="RData" help="A dataframe containing protein ids and protein sequences in an RData file" label="Protein Sequences" />
        <param name="proCodingSeq" type="data" format="RData" help="A dataframe cotaining coding sequences for each protein in an RData file" label="Protein Coding Sequences" />
        <param name="ids" type="data" format="RData" help="A dataframe cotaining IDs for each protein in an RData file" label="Protein IDs" />
        <param name="bamInput" type="data" format="bam" label="BAM file">
            <validator check="bam_index" message="Metadata missing, click the pencil icon in the history item and use the auto-detect feature to correct this issue." type="metadata" />
        </param>
        <!--<param name="baiInput" type="data" format="bam_index" label="BAM Index file">
            <validator check="dbkey" message="Metadata missing, click the pencil icon in the history item and use the auto-detect feature to correct this issue." type="metadata" />
        </param>-->
        <param name="vcfInput" type="data" format="vcf" label="VCF file" />
        <param name="dbsnpInCoding" type="data" format="RData" label="A dataframe containing dbSNP rsids" optional="true" />
        <param name="cosmic" type="data" format="RData" label="A dataframe containing somatic SNPs from COSMIC (human only)" optional="true" />
      </when>
    </conditional>
    <param name="rpkmCutoff" type="float" value="1" min="0" label="Transcript Expression Cutoff (RPKM)" help="If non-zero, if a transcript does not meet this expression cutoff (based on RPKM) then it will not be included in the output database." />
    <param name="outputIndels" type="boolean" truevalue="--outputIndels" falsevalue="" label="Create a variant FASTA for short insertions and deletions" />
    <param name="outputSQLite" type="boolean" truevalue="--outputSQLite" falsevalue="" label="Create SQLite files for mapping proteins to genome and summarizing variant proteins" />
    <param name="outputRData" type="boolean" truevalue="--outputRData" falsevalue="" label="Create RData file of variant protein coding sequences" help="The PSM2SAM tool needs this to map variant proteins to genomic locations" />
  </inputs>
  <outputs>
    <data format="fasta" name="output_rpkm" from_work_dir="output_rpkm.fasta" label="${genome_annotation.bamInput.name.rsplit('.',1)[0]}_rpkm.fasta"/>
    <data format="fasta" name="output_snv" from_work_dir="output_snv.fasta" label="${genome_annotation.bamInput.name.rsplit('.',1)[0]}_snv.fasta"/>
    <data format="fasta" name="output_indel" from_work_dir="output_indel.fasta" label="${genome_annotation.bamInput.name.rsplit('.',1)[0]}_indel.fasta">
      <filter>outputIndels is True</filter>
    </data>
    <data format="rdata" name="output_variant_annotation_rdata" from_work_dir="output.rdata" label="${genome_annotation.bamInput.name.rsplit('.',1)[0]}_variantAnnotation.RData">
      <filter>outputRData is True</filter>
    </data>
    <data format="sqlite" name="output_genomic_mapping_sqlite" from_work_dir="output_genomic_mapping.sqlite" label="${genome_annotation.bamInput.name.rsplit('.',1)[0]}_genomicMapping.sqlite">
      <filter>outputSQLite is True</filter>
    </data>
    <data format="sqlite" name="output_variant_annotation_sqlite" from_work_dir="output_variant_annotation.sqlite" label="${genome_annotation.bamInput.name.rsplit('.',1)[0]}_variantAnnotation.sqlite">
      <filter>outputSQLite is True</filter>
    </data>
  </outputs>
  <tests>
    <test expect_num_outputs="5">
      <param name="bamInput" value="hg19/test1_sort.bam" dbkey="hg19" />
      <param name="vcfInput" value="hg19/test1.vcf" dbkey="hg19" />
      <param name="source" value="history" />
      <param name="exonAnno" value="hg19/exon_anno.RData" />
      <param name="proteinSeq" value="hg19/proseq.RData" />
      <param name="proCodingSeq" value="hg19/procodingseq.RData" />
      <param name="ids" value="hg19/ids.RData" />
      <param name="rpkmCutoff" value="1" />
      <param name="outputIndels" value="" />
      <param name="outputRData" value="--outputRData" />
      <param name="outputSQLite" value="--outputSQLite" />
      <output name="output_rpkm" file="hg19_rpkm1.fasta" />
      <output name="output_snv" file="hg19_snv.fasta" />
      <output name="output_variant_annotation_rdata" file="hg19_variant_annotation.rdata" />
      <output name="output_genomic_mapping_sqlite" file="hg19_genomic_mapping.sqlite" />
      <output name="output_variant_annotation_sqlite" file="hg19_variant_annotation.sqlite" />
    </test>
    <test expect_num_outputs="5">
      <param name="bamInput" value="hg19/test1_sort.bam" dbkey="hg19" />
      <param name="vcfInput" value="hg19/test1.vcf" dbkey="hg19" />
      <param name="source" value="history" />
      <param name="exonAnno" value="hg19/exon_anno.RData" />
      <param name="proteinSeq" value="hg19/proseq.RData" />
      <param name="proCodingSeq" value="hg19/procodingseq.RData" />
      <param name="ids" value="hg19/ids.RData" />
      <param name="dbsnpInCoding" value="hg19/dbsnpinCoding.RData" />
      <param name="cosmic" value="hg19/cosmic.RData" />
      <param name="rpkmCutoff" value="0" />
      <param name="outputIndels" value="--outputIndels" />
      <param name="outputSQLite" value="--outputSQLite" />
      <output name="output_rpkm" file="hg19_rpkm0.fasta" />
      <output name="output_snv" file="hg19_dbsnp_snv.fasta" />
      <output name="output_indel" file="hg19_indel.fasta" />
      <output name="output_variant_annotation_rdata" file="hg19_dbsnp_variant_annotation.rdata" />
      <output name="output_variant_annotation_sqlite" file="hg19_dbsnp_variant_annotation.sqlite" />
    </test>
    <test expect_num_outputs="3">
      <param name="bamInput" value="hg19/test1_sort.bam" dbkey="hg19" />
      <param name="vcfInput" value="hg19/test1.vcf" dbkey="hg19" />
      <param name="source" value="history" />
      <param name="exonAnno" value="hg19/exon_anno.RData" />
      <param name="proteinSeq" value="hg19/proseq.RData" />
      <param name="proCodingSeq" value="hg19/procodingseq.RData" />
      <param name="ids" value="hg19/ids.RData" />
      <param name="rpkmCutoff" value="1000" />
      <param name="outputIndels" value="--outputIndels" />
      <output name="output_rpkm" file="hg19_rpkm1000.fasta" />
      <output name="output_snv" file="hg19_snv.fasta" />
      <output name="output_indel" file="hg19_indel.fasta" />
    </test>
  </tests>
  <help>
**Description**

Generate custom protein FASTAs from exosome or transcriptome data.

The reference protein set can be filtered by transcript expression level (RPKM calculated from a BAM file),
and variant protein forms can be predicted based on variant calls (SNPs and INDELs reported in a VCF file).


**Annotations**
CustomProDB depends on a set of annotation files (in RData format) to create reference and variant protein sequences.
Galaxy administrators can use the customProDB data manager to create these annotations to make them available for users.
</help>
  <citations>
    <citation type="doi">10.1093/bioinformatics/btt543</citation>
    <citation type="bibtex">@misc{toolsGalaxyP, author = {Chambers MC, et al.}, title = {Galaxy Proteomics Tools}, publisher = {GitHub}, journal = {GitHub repository},
                                  year = {2017}, url = {https://github.com/galaxyproteomics/tools-galaxyp}}</citation> <!-- TODO: fix substitution of commit ", commit = 
{$sha1$}" -->
  </citations>
</tool>