<tool id="exomedepth" name="ExomeDepth" version="1.1.0">
    <description>Calls copy number variants (CNVs) from targeted sequence data</description>
    <requirements>
        <requirement type="package" version="1.1.10">r-exomedepth</requirement>
    </requirements>
    <version_command><![CDATA[
echo $(R --version | grep version | grep -v GNU)", ExomeDepth version" $(R --vanilla --slave -e "library(ExomeDepth); cat(sessionInfo()\$otherPkgs\$ExomeDepth\$Version)")
    ]]></version_command>
    <command detect_errors="exit_code"><![CDATA[
    Rscript '${__tool_directory__}/exomedepth.R' '$args_file'
    ]]></command>
    <configfiles>
        <configfile name="args_file"><![CDATA[
target=$targetFile
test_vs_ref=$test_vs_ref
#for $i in $inputs
bam=${i.input}
bam_bai=${i.input.metadata.bam_index}
#if str($i.label.value) != "":
bam_label=${$i.label.value}
#else
bam_label=${i.input.dataset.name}
#end if
#end for
trans_prob=$transition_probability
output=$output
        ]]></configfile>
    </configfiles>
    <inputs>
        <param name="targetFile" type="data" format="bed" label="Target regions (BED)">
            <validator type="unspecified_build" />
        </param>
        <param name="test_vs_ref" type="boolean" truevalue="TRUE" falsevalue="FALSE" checked="false" label="Call CNVs using 1st sample as test" help="If checked, the tool will call CNVs in the first sample vs all the others. If unchecked, an all vs all CNV call will be performed" />
        <repeat name="inputs" title="BAM" min="2" help="Need to add more files? Use controls below.">
            <param name="input" type="data" format="bam" label="BAM file">
                <options>
                    <filter type="data_meta" ref="targetFile" key="dbkey"/>
                </options>
            </param>
            <param name="label" type="text" size="30" value="" label="Label" help="Label to use in the output. If not given, the dataset name will be used instead">
                <validator type="regex" message="Spaces are not allowed">^\S*$</validator>
            </param>  
        </repeat>
        <param name="transition_probability" size="10" type="float" value="0.0001" label="Transition probability" help="Transition probability of the hidden Markov Chain from the normal copy number state to either a deletion or a duplication. The default value (0.0001) expects approximately 20 CNVs genome-wide" />
    </inputs>
    <outputs>
        <data name="output" format="tabular" label="${tool.name} on ${on_string}" />
    </outputs>
    <tests>
        <test>
            <param name="targetFile" value="CNV_TruSeq_Chr2.bed" dbkey="hg19" ftype="bed"/>
            <param name="test_vs_ref" value="True"/>
            <repeat name="inputs">
                <param name="input" value="CNV_case_small.bam"/>
            </repeat>
            <repeat name="inputs">
                <param name="input" value="CNV_control_small.bam"/>
            </repeat>
            <param name="transition_probability" value="0.5"/>
            <output name="output">
                <assert_contents>
                    <has_text text="chr2" />
                    <has_text text="97890544" />
                    <has_text text="97890616" />
                    <has_text text="deletion" />
                    <has_text text="CNV_case_small" />
                </assert_contents>
            </output>
        </test>
    </tests>
    <help><![CDATA[
.. class:: warningmark

**Warning about counts for chromosome X**

Calling CNVs on the X chromosome can create issues if the exome sample of interest and the reference exome
samples it is being compared to are not gender matched.
Make sure that the genders are matched properly (i.e. do not use male as a reference for female
samples and vice versa).

**What it does**

This tool uses ExomeDepth to call copy number variants (CNVs) from targeted sequence data.

**Output format**

=========== ========================
Column      Description
----------- ------------------------
chr         Chromosome
start       Start of CNV region
end         End of CNV region
type        CNV type (deletion, duplication)
sample      Name of the sample with CNV
corr        Correlation between reference and test counts. To get meaningful result, this correlation should really be above 0.97. If this is not the case, consider the output of ExomeDepth as less reliable (i.e. most likely a high false positive rate) 
nexons      Number of target regions covered by the CNV
BF          Bayes factor. It quantifies the statistical support for each CNV. It is in fact the log10 of the likelihood ratio of data for the CNV call divided by the null (normal copy number). The higher that number, the more confident one can be about the presence of a CNV. While it is difficult to give an ideal threshold, and for short exons the Bayes Factor are bound to be unconvincing, the most obvious large calls should be easily flagged by ranking them according to this quantity
reads.ratio Observed/expected reads ratio
=========== ========================


**What ExomeDepth does and does not do**

ExomeDepth uses read depth data to call CNVs from exome sequencing experiments. A key idea is that the test 
exome should be compared to a matched aggregate reference set. This aggregate reference set should combine 
exomes from the same batch and it should also be optimized for each exome. It will certainly differ from one exome 
to the next.

Importantly, ExomeDepth assumes that the CNV of interest is absent from the aggregate reference set. Hence 
related individuals should be excluded from the aggregate reference. It also means that ExomeDepth can miss 
common CNVs, if the call is also present in the aggregate reference. ExomeDepth is really suited to detect rare 
CNV calls (typically for rare Mendelian disorder analysis).

The ideas used in this package are of course not specific to exome sequencing and could be applied to other 
targeted sequencing datasets, as long as they contain a sufficiently large number of exons to estimate the parameters 
(at least 20 genes, say, but probably more would be useful). Also note that PCR based enrichment studies are often 
not well suited for this type of read depth analysis. The reason is that as the number of cycles is often set to a high 
number in order to equalize the representation of each amplicon, which can discard the CNV information.
    ]]></help>
    <citations>
        <citation type="doi">10.1093/bioinformatics/btu135</citation>
        <citation type="doi">10.1093/bioinformatics/bts526</citation>
    </citations>
</tool>