<tool id="diffbind" name="DiffBind" version="2.6.6.2">
    <description> differential binding analysis of ChIP-Seq peak data</description>
    <requirements>
        <requirement type="package" version="2.6.6">bioconductor-diffbind</requirement>
        <requirement type="package" version="1.20.0">r-getopt</requirement>
        <requirement type="package" version="0.2.15">r-rjson</requirement>
    </requirements>
    <stdio>
        <regex match="Execution halted"
           source="both"
           level="fatal"
           description="Execution halted." />
        <regex match="Input-Error 01"
           source="both"
           level="fatal"
           description="Error in your input parameters: Make sure you only apply factors to selected samples." />
        <regex match="Error in"
           source="both"
           level="fatal"
           description="An undefined error occured, please check your input carefully and contact your administrator." />
    </stdio>
    <version_command><![CDATA[
echo $(R --version | grep version | grep -v GNU)", DiffBind version" $(R --vanilla --slave -e "library(DiffBind); cat(sessionInfo()\$otherPkgs\$DiffBind\$Version)" 2> /dev/null | grep -v -i "WARNING: ")", rjson version" $(R --vanilla --slave -e "library(rjson); cat(sessionInfo()\$otherPkgs\$rjson\$Version)" 2> /dev/null | grep -v -i "WARNING: ")
    ]]></version_command>
    <command><![CDATA[
#import re
#import json

## Adapted from DESeq2 wrapper
#set $temp_factor_names = list()
#set $temp_factor = list()

#for $g in $rep_group:

    #set $peak_files = list()
    #set $bam_files = list()
    #set $bam_controls = list()

    #for $file in $g.peaks:
        #set $file_name = re.sub('[^\w\-\s]', '_', str($file.element_identifier))
        ln -s '${file}' ${g.groupName}-${file_name}-peaks.bed &&
        $peak_files.append(str($g.groupName) + '-' + $file_name + '-peaks.bed')
    #end for

    #for $bam in $g.bamreads:
        #set $bam_name = re.sub('[^\w\-\s]', '_', str($bam.element_identifier))
        ln -s '${bam}' ${bam_name}-bamreads.bam &&
        ln -s ${bam.metadata.bam_index} ${bam_name}-bamreads.bai &&
        $bam_files.append($bam_name + '-bamreads.bam')
    #end for

    $temp_factor.append( {str($g.groupName): $peak_files} )
    $temp_factor.append( {str($g.groupName): $bam_files} )

    #if str( $g.bamcontrol ) != 'None':
        #for $ctrl in $g.bamcontrol:
            #set $ctrl_name = re.sub('[^\w\-\s]', '_', str($ctrl.element_identifier))
            ln -s '${ctrl}' ${g.groupName}-${ctrl_name}-bamcontrol.bam &&
            ln -s ${ctrl.metadata.bam_index} ${g.groupName}-${ctrl_name}-bamcontrol.bai &&
            $bam_controls.append(str($g.groupName) + '-' + $ctrl_name + '-bamcontrol.bam')
        #end for
        $temp_factor.append( {str($g.groupName): $bam_controls} )
    #end if

#end for

$temp_factor.reverse()
$temp_factor_names.append(["Condition", $temp_factor])


Rscript '$__tool_directory__/diffbind.R'

    -i '#echo json.dumps(temp_factor_names)#'
    -o '$outfile'
    -t $th
    -f $out.format
    -p '$plots'

    #if $scorecol:
        -n "$scorecol"
    #end if
    #if $lowerbetter:
        -l "$lowerbetter"
    #end if
    #if $summits:
        -s "$summits"
    #end if

    #if $out.binding_matrix:
        -b
    #end if

    #if $out.rdata:
        -r
    #end if

    #if $out.analysis_info:
        -a
    #end if

    #if $out.rscript:
        && cp '$__tool_directory__/diffbind.R' '$rscript'
    #end if
]]>
    </command>
    <inputs>
        <repeat name="rep_group" title="Group" min="2" max="2" default="2">
            <param name="groupName" type="text" label="Name"
            help="Name for the Group that the peak and BAM files belong to e.g. Resistant/Responsive (two Groups in total must be specified for DiffBind). NOTE: Please only use letters, numbers or underscores.">
                <sanitizer>
                    <valid initial="string.letters,string.digits"><add value="_" /></valid>
                </sanitizer>
                <validator type="empty_field" />
            </param>
            <param name="peaks" type="data" format="bed" multiple="true" label="Peak files" help="Result of your Peak calling experiment"/>
            <param name="bamreads" type="data" format="bam" multiple="true" label="Read BAM file" help="Specify the Read BAM file used in the Peak calling."/>
            <param name="bamcontrol" type="data" format="bam" multiple="true" optional="True" label="Control BAM file" help="If specifying a control BAM file, all samples are required to specify one, see Help section below."/>
        </repeat>

        <param name="scorecol" type="integer" min="0" value="8" label="Score Column" help="Column in peak files that contains peak scores. Default: 8 (narrowPeak)">
            <sanitizer>
                <valid initial="string.digits"/>
            </sanitizer>
        </param>
        <param name="lowerbetter" type="boolean" truevalue="True" falsevalue="" checked="False" label="Lower score is better?" help="DiffBind by default assumes that a higher score indicates a better peak, for example narrowPeaks -log10pvalue. If this is not the case, for example if the score is a p-value or FDR, set this option to Yes. Default: No" />
        <param name="summits" type="integer" min="0" optional="True" label="Summits" help="Extend peaks Nbp up- and downstream of the summit. For punctate peaks it is advisable to extend (e.g. 250bp), see the DiffBind User Guide">
            <sanitizer>
                <valid initial="string.digits"/>
            </sanitizer>
        </param>
        <param name="th" type="float" value="0.05" min="0" max="1" label="FDR Threshold" help="Significance threshold; all sites with FDR less than or equal to this value will be included in the output. A value of 1 will output all binding sites. Default: 0.05"/>

        <!-- Output Options -->
        <section name="out" expanded="false" title="Output Options">
            <param name="format" type="select" label="Output Format">
                <option value="bed">BED</option>
                <option value="gff">GFF</option>
                <option value="wig">WIG</option>
            </param>
            <param name="pdf" type="boolean" truevalue="True" falsevalue="" checked="False" label="Visualising the analysis results" help="output an additional PDF file" />
            <param name="binding_matrix" type="boolean" truevalue="True" falsevalue="" checked="False" label="Output binding affinity matrix?" help="Output a table of the binding scores" />
            <param name="rdata" type="boolean" truevalue="True" falsevalue="" checked="False" label="Output RData file?" help="Output all the data used by R to construct the plots and tables, can be loaded into R. Default: No"/>
            <param name="rscript" type="boolean" truevalue="True" falsevalue="False" checked="False" label="Output Rscript?" help="If this option is set to Yes, the Rscript used will be provided as a text file in the output. Default: No"/>
            <param name="analysis_info" type="boolean" truevalue="True" falsevalue="False" checked="False" label="Output analysis info?" help="If this option is set to Yes, information from the dba.count and dba.analyze commmands will be output in a text file. Default: No"/>
        </section>
    </inputs>

    <outputs>
        <data name="outfile" format="tabular" label="${tool.name} on ${on_string}: Differentially bound sites" />
        <data name="plots" format="pdf" label="${tool.name} on ${on_string}: Plots">
            <filter>out['pdf']</filter>
        </data>
        <data name="binding_matrix" format="tabular" from_work_dir="bmatrix.tab" label="${tool.name} on ${on_string}: Binding matrix">
            <filter>out['binding_matrix']</filter>
        </data>
        <data name="rdata" format="rdata" from_work_dir="DiffBind_analysis.RData" label="${tool.name} on ${on_string}: RData file">
            <filter>out['rdata']</filter>
        </data>
        <data name="rscript" format="txt" label="${tool.name} on ${on_string}: Rscript">
            <filter>out['rscript']</filter>
        </data>
        <data name="analysis_info" format="txt" from_work_dir="DiffBind_analysis_info.txt" label="${tool.name} on ${on_string}: Analysis info">
            <filter>out['analysis_info']</filter>
        </data>
    </outputs>

    <tests>
        <test expect_num_outputs="6">
            <repeat name="rep_group">
                <param name="groupName" value="Resistant"/>
                <param name="peaks" value="BT474_ER_1.bed.gz,BT474_ER_2.bed.gz"/>
                <param name="bamreads" ftype="bam" value="BT474_ER_1.bam,BT474_ER_2.bam" />
            </repeat>
            <repeat name="rep_group">
                <param name="groupName" value="Responsive"/>
                <param name="peaks" value="MCF7_ER_1.bed.gz,MCF7_ER_2.bed.gz"/>
                <param name="bamreads" ftype="bam" value="MCF7_ER_1.bam,MCF7_ER_2.bam" />
            </repeat>
            <param name="scorecol" value="5" />
            <param name="pdf" value="True" />
            <param name="binding_matrix" value="True" />
            <param name="rdata" value="True" />
            <param name="rscript" value="True"/>
            <param name="analysis_info" value="True"/>
            <output name="outfile" value="out_diffbind.tab" />
            <output name="plots" value="out_plots.pdf" compare="sim_size" />
            <output name="binding_matrix" value="out_binding.matrix" />
            <output name="rdata" value="DiffBind_analysis.RData" compare="sim_size"/>
            <output name="rscript" value="out_rscript.txt"/>
            <output name="analysis_info" value="out_analysis_info.txt" compare="sim_size" >
                <assert_contents>
                    <has_text text="SessionInfo"/>
                </assert_contents>
            </output>
        </test>
    </tests>
    <help><![CDATA[

.. class:: infomark

**What it does**

DiffBind_ is a `Bioconductor package`_ that provides functions for processing ChIP-Seq data enriched for genomic loci where specific
protein/DNA binding occurs, including peak sets identified by ChIP-Seq peak callers and
aligned sequence read datasets. It is designed to work with multiple peak sets simultaneously,
representing different ChIP experiments (antibodies, transcription factor and/or histone
marks, experimental conditions, replicates) as well as managing the results of multiple peak
callers.

The primary emphasis of DiffBind is on identifying sites that are differentially bound
between two sample groups. It includes functions to support the processing of peak sets,
including overlapping and merging peak sets, counting sequencing reads overlapping intervals
in peak sets, and identifying statistically significantly differentially bound sites based on
evidence of binding affinity (measured by differences in read densities). To this end it uses
statistical routines developed in an RNA-Seq context (primarily the Bioconductor packages
edgeR and DESeq2). Additionally, the package builds on Rgraphics routines to provide a
set of standardized plots to aid in binding analysis.

The `DiffBind User Guide`_ includes a brief overview of the processing flow, followed by four sections of
examples: the first focusing on the core task of obtaining differentially bound sites based on
affinity data, the second working through the main plotting routines, the third discussing the
use of a blocking factor, and the fourth revisiting occupancy data (peak calls) in more detail,
as well as comparing the results of an occupancy-based analysis with an affinity-based one.
Finally, certain technical aspects of the how these analyses are accomplished are detailed.

Note this DiffBind tool requires a minimum of four samples (two groups with two replicates each).

-----

**Inputs**

DiffBind works primarily with peaksets, which are sets of genomic intervals representing
candidate protein binding sites. Each interval consists of a chromosome, a start and end
position, and usually a score of some type indicating confidence in, or strength of, the peak.
Associated with each peakset are metadata relating to the experiment from which the peakset
was derived. Additionally, files containing mapped sequencing reads (BAM files) need to
be associated with each peakset (one for the ChIP data, and optionally another representing
a control sample)

**Groups**

You have to specify the name of the Group and the peak and BAM files for the two Groups you want to compare (e.g Resistant and Responsive) in the tool form above.

Example:

    ============= =============
     **Sample**   **Group**
    ------------- -------------
    BT4741        Resistant
    BT4742        Resistant
    MCF71         Responsive
    MCF72         Responsive
    ============= =============


**Peak files**

Result of your Peak calling experiment in bed format, one file for each sample is required. The peak caller, format and score column can be specified in the tool form above. The default settings expect narrowPeak bed format, which has the score in the 8th column (-log10pvalue), and can be output from MACS2.

Example:

    ======= ======= ======= =============== ==============
    1          2      3          4           **5 (Score)**
    ======= ======= ======= =============== ==============
    chr18   215562  216063  peak_16037      56.11
    chr18   311530  312105  peak_16038      222.49
    chr18   356656  357315  peak_16039      92.06
    chr18   371110  372092  peak_16040      123.86
    chr18   395116  396464  peak_16041      1545.39
    chr18   399014  400382  peak_16042      1835.19
    chr18   499134  500200  peak_16043      748.32
    chr18   503518  504552  peak_16044      818.30
    chr18   531672  532274  peak_16045      159.30
    chr18   568326  569282  peak_16046      601.11
    ======= ======= ======= =============== ==============

* BAM file which contains the mapped sequencing reads associated with each peakset, one file for each sample is required.
* Optional: Control BAM file representing a control dataset. If used, has to be specified for all samples. Note that the DiffBind authors say control reads are best utilized prior to running DiffBind, at the peak calling stage (e.g. with MACS2) and in blacklists, see this `Bioconductor post`_.

-----

**Outputs**

This tool outputs

    * a table of differentially bound sites

Optionally, under **Output Options** you can choose to output

    * a PDF of plots (Heatmap, PCA, MA, Volcano, Boxplots)
    * a binding affinity matrix
    * the R script used by this tool
    * an RData file of the R objects generated
    * a text file with information on the analysis (number of Intervals, FriP scores, method used)

**Differentially Bound Sites**

Example:

    ========  ======  ======  =====  ======  =====  ===============  ==============  ======  ========  ========
    seqnames  start   end     width  strand  Conc   Conc_Responsive  Conc_Resistant  Fold    p.value   **FDR**
    ========  ======  ======  =====  ======  =====  ===============  ==============  ======  ========  ========
    chr18     394600  396513  1914    \*     7.15   5.55             7.89            -2.35   7.06e-24  9.84e-21
    chr18     111567  112005  439     \*     5.71   6.53             3.63            2.89    1.27e-08  8.88e-06
    chr18     346464  347342  879     \*     5      5.77             3.24            2.52    6.51e-06  0.00303
    chr18     399014  400382  1369    \*     7.62   7                8.05            -1.04   1.04e-05  0.00364
    chr18     371110  372102  993     \*     4.63   3.07             5.36            -2.3    8.1e-05   0.0226
    ========  ======  ======  =====  ======  =====  ===============  ==============  ======  ========  ========

Columns contain the following data:

* **seqnames**: Chromosome name
* **start**: Start position of site
* **end**: End position of site
* **width**: Length of site
* **strand**: Strand
* **Conc**: Mean read concentration over all the samples (the default calculation uses log2 normalized ChIP read counts with control read counts subtracted)
* **Responsive**: Mean concentration over the first (e.g. Responsive) group
* **Resistant**: Mean concentration over second (e.g. Resistant) group
* **Fold**: Fold shows the difference in mean concentrations between the two groups (e.g. Responsive - Resistant), with a positive value indicating increased binding affinity in the first group and a negative value indicating increased binding affinity in the second group.
* **p.value**: P-value confidence measure for identifying these sites as differentially bound
* **FDR**: a multiple testing corrected FDR p-value


**Binding Affinity Matrix**

The final result of counting is a binding affinity matrix containing a (normalized) read count for each sample at every potential binding site. With this matrix, the samples can be re-clustered using affinity, rather than occupancy, data. The binding affinity matrix can be used for QC plotting as well as for subsequent
differential analysis.

Example:

    =====  ======  ======  =========  =========  ==========  ==========
    CHR    START   END     MCF7_ER_1  MCF7_ER_2  BT474_ER_1  BT474_ER_2
    =====  ======  ======  =========  =========  ==========  ==========
    chr18  111567  112005  137.6152   59.87837   29.41393    19.95945
    chr18  189223  189652  19.95945   12.60597   11.55547    23.11095
    chr18  215232  216063  11.55547   15.75746   31.51493    72.48434
    chr18  311530  312172  17.85846   11.55547   54.62588    43.07040
    chr18  346464  347342  75.63583   40.96941   21.00995    16.80796
    chr18  356560  357362  11.55547   14.70696   57.77737    53.57538
    chr18  371110  372102  8.403982   9.454479   81.93882    82.98932
    chr18  394600  396513  56.72687   43.07040   510.5419    438.0575
    chr18  399014  400382  156.5241   117.6557   558.8648    496.8854
    chr18  498906  500200  767.9138   278.3819   196.4430    181.7361
    =====  ======  ======  =========  =========  ==========  ==========

-----

**More Information**

Generally, processing data with DiffBind involves five phases:

 #. Reading in peaksets
 #. Occupancy analysis
 #. Counting reads
 #. Differential binding affinity analysis
 #. Plotting and reporting


**Reading in peaksets**:

The first step is to read in a set of peaksets and associated
metadata. Peaksets are derived either from ChIP-Seq peak callers, such as **MACS2**, or using some other criterion (e.g. genomic windows, or all the promoter regions
in a genome).  A single experiment can have more than
one associated peakset; e.g. if multiple peak callers are used for comparison purposes
each sample would have more than one line in the sample sheet. Once the peaksets
are read in, a merging function finds all overlapping peaks and derives a single set of
unique genomic intervals covering all the supplied peaks (a consensus peakset for the
experiment).

**Occupancy analysis**:

Peaksets, especially those generated by peak callers, provide
an insight into the potential occupancy of the protein being ChIPed for at specific
genomic loci. After the peaksets have been loaded, it can be useful to perform some
exploratory plotting to determine how these occupancy maps agree with each other,
e.g. between experimental replicates (re-doing the ChIP under the same conditions),
between different peak callers on the same experiment, and within groups of samples
representing a common experimental condition. DiffBind provides functions to enable
overlaps to be examined, as well as functions to determine how well similar samples
cluster together. Beyond quality control, the product of an occupancy analysis may be
a consensus peakset, representing an overall set of candidate binding sites to be used
in further analysis.

**Counting reads**:

Once a consensus peakset has been derived, DiffBind can use the
supplied sequence read files to count how many reads overlap each interval for each
unique sample. The peaks in the consensus peakset may be re-centered and trimmed
based on calculating their summits (point of greatest read overlap) in order to provide
more standardized peak intervals. The final result of counting is a binding affinity matrix
containing a (normalized) read count for each sample at every potential binding site.
With this matrix, the samples can be re-clustered using affinity, rather than occupancy,
data. The binding affinity matrix is used for QC plotting as well as for subsequent
differential analysis.

**Differential binding affinity analysis**:

The core functionality of DiffBind is the
differential binding affinity analysis, which enables binding sites to be identified that
are statistically significantly differentially bound between sample groups. To accomplish
this, first a contrast (or contrasts) is established, dividing the samples into groups to
be compared. Next the core analysis routines are executed, by default using DESeq2.
This will assign a p-value and FDR to each candidate binding site indicating confidence
that they are differentially bound.

**Plotting and reporting**:

Once one or more contrasts have been run, DiffBind provides
a number of functions for reporting and plotting the results. MA plots give an
overview of the results of the analysis, while correlation heatmaps and PCA plots show
how the groups cluster based on differentially bound sites. Boxplots show the distribution
of reads within differentially bound sites corresponding to whether they gain or
lose affinity between the two sample groups. A reporting mechanism enables differentially
bound sites to be extracted for further processing, such as annotation, motif, and
pathway analyses.

-----

**References**

DiffBind Authors:  Rory Stark, Gordon Brown (2011)
Wrapper authors: Bjoern Gruening, Pavankumar Videm

.. _DiffBind: https://bioconductor.org/packages/release/bioc/html/DiffBind.html
.. _`Bioconductor package`: https://bioconductor.org/packages/release/bioc/html/DiffBind.html
.. _`DiffBind User Guide`: https://bioconductor.org/packages/release/bioc/vignettes/DiffBind/inst/doc/DiffBind.pdf
.. _`Bioconductor post`: https://support.bioconductor.org/p/69924/

]]>
    </help>
    <citations>
        <citation type="doi">doi:10.1038/nature10730</citation>
    </citations>
</tool>