<tool id="tgsgapcloser" name="TGS-GapCloser" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="20.01">
    <description>fills the N-gap of error-prone long reads</description>
    <macros>
        <import>macros.xml</import>
    </macros>
    <expand macro="requirements" />
    <expand macro="xrefs"/>
    <version_command>tgsgapcloser --version</version_command>
    <command detect_errors="exit_code"><![CDATA[
        PILON=\$(which pilon)
        PILON_JAR=\$(readlink -f \$PILON).jar
        && tgsgapcloser    
        --scaff $scaff 
        --reads $reads
        --output output
        #if $error_conditional.error_options == 'pilon'
            --pilon \$PILON_JAR
            --ngs $error_conditional.ngs
            --samtools `which samtools`
            --java `which java`
            --pilon_mem \${GALAXY_MEMORY_MB:-4096}M
        #elif $error_conditional.error_options == 'racon'
            --racon `which racon`
        #else
            --ne
        #end if
        --tgstype $tgstype_conditional.tgstype
        --min_idy $tgstype_conditional.min_idy
        --min_match $tgstype_conditional.min_match
        --chunk $chunk
        $g_check
        --thread  \${GALAXY_SLOTS:-16}
    ]]></command>
    <inputs>
        <param argument="--scaff" type="data" format="fasta" multiple="true" label="Scaffold file" />
        <param argument="--reads" type="data" format="fasta" multiple="true" label="Input reads" />
        <conditional name="error_conditional">
            <param name="error_options" type="select" label="Error correction">
                <option value="ne">Do not error correct</option>
                <option value="racon">Racon</option>
                <option value="pilon">Pilon</option>
            </param>
            <when value="ne"/>
            <when value="racon">
                <param argument="--r_round" type="integer" min="0" max="10" value="1" label="Number of Racon error-correction rounds" 
                    help="Although multiple rounds of racon can increase the quality of an assembly there are indications that it also 
                        fragments the assembly and may decrease quality by removing structural variants and SNPs. Published assembly workflows 
                        differ in the number of rounds but rarely apply more than 4 rounds of racon" />
            </when>
            <when value="pilon">
                <param argument="--ngs" type="data" format="fastq,fastq.gz" label="Illumina reads" 
                    help="Pilon can utilize Illumina short reads mapped to the draft assembly to 
                        improve the local accuracy of the sequence by correcting sequence errors, 
                        fixing misassemblies, and filling gaps"/>
                <param argument="--p_round" type="integer" min="0" max="10" value="3" label="Number of Pilon error-correction rounds" />
            </when>
        </conditional>
        <conditional name="tgstype_conditional">
            <param argument="--tgstype" type="select" label="Type of third generation reads">
                <option value="ont" selected="true">Oxford Nanopore Technologies (ONT)</option>
                <option value="pb">PacBio (pb)</option>
            </param>
            <when value="ont">
                <param argument="--min_idy" type="float" min="0" max="1" value="0.3" label="Minimum identity for filtering candidate sequences"/>
                <param argument="--min_match" type="integer" min="0" max="1000" value="300" label="Minimum matched length for filtering candidate sequences"/>
            </when>
            <when value="pb">
                <param argument="--min_idy" type="float" min="0" max="1" value="0.2" label="Minimum identity for filtering candidate sequences"/>
                <param argument="--min_match" type="integer" min="0" max="1000" value="200" label="Minimum matched length for filtering candidate sequences"/>
            </when>
        </conditional>
        <param argument="--chunk" type="integer" min="0" max="20" value="3" label="Chunks for error correction" help="Split candidates into # of chunks to separately correct errors" />
        <param argument="--g_check" type="boolean" truevalue="--g_check" falsevalue="" label="Gap-size diff check"/>
        <param name="output_options" type="select" multiple="true" optional="true" display="checkboxes" label="Output files">
            <option value="log_file">General log file</option>
            <option value="gapfill_log">Gapfill log file</option>
        </param>
    </inputs>
    <outputs>
        <data name="log" format="txt" from_work_dir="output.fill.log" label="${tool.name} on ${on_string}: log">
            <filter>output_options and 'log_file' in output_options</filter>
        </data>
        <data name="final_assembly" format="fasta" from_work_dir="output.scaff_seqs" label="${tool.name} on ${on_string}: final assembly"/>

        <data name="fill_details" format="txt" from_work_dir="output.gap_fill_detail" label="${tool.name} on ${on_string}: gap fill details">
            <filter>output_options and 'gapfill_log' in output_options</filter>
        </data>
    </outputs>
    <tests>
        <!--Test 01: no correction-->
        <test expect_num_outputs="3">
            <param name="scaff" value="scaffold.fasta" ftype="fasta"/>
            <param name="reads" value="ont_reads.fasta" ftype="fasta"/>
            <conditional name="error_conditional">
                <param name="error_options" value="ne"/>
            </conditional>
            <conditional name="tgstype_conditional">
                <param name="tgstype" value="ont"/>
                <param name="min_idy" value="0.3"/>
                <param name="min_match" value="300"/>
            </conditional>
            <param name="chunk" value="3"/>
            <param name="g_check" value="false"/>
            <param name="output_options" value="log_file,gapfill_log"/>
            <output name="final_assembly" file="test_01_final_assembly.fasta" ftype="fasta"/>
            <output name="log" ftype="txt">
                <assert_contents>
                    <has_text text="TGSGapCloser start now"/>
                    <has_text text="the one read provide filler choose count freq for a gap"/>
                </assert_contents>
            </output>
            <output name="fill_details" file="test_01_gapfill.log" ftype="txt"/>
        </test>
        <!--Test 02: correction with racon-->
        <test expect_num_outputs="1">
            <param name="scaff" value="scaffold.fasta" ftype="fasta"/>
            <param name="reads" value="ont_reads.fasta" ftype="fasta"/>
            <conditional name="error_conditional">
                <param name="error_options" value="racon"/>
                <param name="r_round" value="2"/>
            </conditional>
            <conditional name="tgstype_conditional">
                <param name="tgstype" value="ont"/>
                <param name="min_idy" value="0.3"/>
                <param name="min_match" value="300"/>
            </conditional>
            <param name="chunk" value="3"/>
            <param name="g_check" value="false"/>
            <output name="final_assembly" file="test_02_final_assembly.fasta" ftype="fasta"/>
        </test>
        <!--Test 03: correction pilon-->
        <test expect_num_outputs="1">
            <param name="scaff" value="scaffold.fasta" ftype="fasta"/>
            <param name="reads" value="ont_reads.fasta" ftype="fasta"/>
            <conditional name="error_conditional">
                <param name="error_options" value="pilon"/>
                <param name="ngs" value="ngs_reads.fastq" ftype="fastq"/>
                <param name="r_round" value="1"/>
            </conditional>
            <conditional name="tgstype_conditional">
                <param name="tgstype" value="ont"/>
                <param name="min_idy" value="0"/>
                <param name="min_match" value="0"/>
            </conditional>
            <param name="chunk" value="1"/>
            <param name="g_check" value="true"/>
            <output name="final_assembly" file="test_03_final_assembly.fasta" ftype="fasta"/>
        </test>
        <!--Test 04: correction with racon and pacbio-->
        <test expect_num_outputs="1">
            <param name="scaff" value="scaffold.fasta" ftype="fasta"/>
            <param name="reads" value="ont_reads.fasta" ftype="fasta"/>
            <conditional name="error_conditional">
                <param name="error_options" value="racon"/>
                <param name="r_round" value="2"/>
            </conditional>
            <conditional name="tgstype_conditional">
                <param name="tgstype" value="pacbio"/>
                <param name="min_idy" value="0.2"/>
                <param name="min_match" value="200"/>
            </conditional>
            <param name="chunk" value="2"/>
            <param name="g_check" value="false"/>
            <output name="final_assembly" file="test_04_final_assembly.fasta" ftype="fasta"/>
        </test>
        <!--Test 05: fastq.gz files-->
        <test expect_num_outputs="1">
            <param name="scaff" value="scaffold.fasta" ftype="fasta"/>
            <param name="reads" value="ont_reads.fasta" ftype="fasta"/>
            <conditional name="error_conditional">
                <param name="error_options" value="pilon"/>
                <param name="ngs" value="ngs_reads.fastq.gz" ftype="fastq.gz"/>
                <param name="r_round" value="1"/>
            </conditional>
            <conditional name="tgstype_conditional">
                <param name="tgstype" value="ont"/>
                <param name="min_idy" value="0"/>
                <param name="min_match" value="0"/>
            </conditional>
            <param name="chunk" value="1"/>
            <param name="g_check" value="false"/>
            <output name="final_assembly" file="test_05_final_assembly.fasta" ftype="fasta"/>
        </test>
    </tests>
    <help><![CDATA[

.. class:: infomark

**Purpose**

TGS-GapCloser is a gap-closing software tool that uses error-prone long reads generated by third-generation-sequence techniques (Pacbio,
Oxford Nanopore, etc.) or preassembled contigs to fill N-gap in the genome assembly. This tool can close gaps in large genome assemblies
using raw long reads quickly and cost-effectively. The final assemblies generated by TGS-GapCloser have improved contiguity and 
completeness while maintaining high accuracy.

----

.. class:: infomark

**Quick usage**

Input reads can only be in FASTA format. Both raw reads and pre-error-corrected reads are acceptable as input. If only raw long reads are 
provided, it polishes raw TGS reads by calling Racon.If additional NGS short reads are available, it polishes raw TGS reads by calling Pilon.

----

.. class:: infomark

**Gap fill details**


Format of a detailed information of gap fill report:

- Each scaffold name is followed by its data lines.
- A data line consists of 3 or 5 columns and describes the source of each segment in the final sequence:
- Column 1 is the segment's first bp position in the final sequence.
- Column 2 is the segment's last bp position in the final sequence.
- Column 3 is the segment's type , 'S' , 'N' or 'F'.
- 'S' means this segment is a segment of the input sequence and this line includes other two more columns:
- Column 4 is the segment's first bp position in the input sequence.
- Column 5 is the segment's last bp position in the input sequence.
- 'N' means this segment is a N area.
- 'F' means this segment is a filled sequence from TGS reads.

----

.. class:: infomark

**Algorithm and implementation of TGS-GapCloser**

This is a brief description of the TGS-GapCloser algorithm. Please refer to the manuscript for more detailed information.

TGS-GapCloser is coded in the C++ programing language (requires GCC 4.4+). It uses minimap2 to obtain alignments, and Pilon (requires Java runtime 1.7+)
or Racon (requires GCC 4.8+) to correct candidate fragments. The algorithm automatically identifies gaps and tries to find the best matched long-read 
fragments to close gaps or merge adjacent scaftigs. To accelerate the gap closure without losing efficiency and accuracy, TGS-GapCloser only selects a 
limited number of fragmented long reads as candidates for subsequent error correction and competition.

TGS-GapCloser can accept as input any type of TGS long reads or other pre-assembled contigs to fill gaps in a draft assembly in the 4 steps :
	(i) Identification of gap regions in the draft assembly; 
	(ii) Acquisition of candidates from the alignments of long reads against gaps; 
	(iii) Base-level error correction of alternative sub-long reads; and 
	(iv) Gap closure using the error-corrected candidates with the highest score for each gap or linkage of the neighboring scaftigs with overlaps.

  ]]></help>
    <expand macro="citations" />
</tool>