<tool id="retrieve_fasta_from_NCBI" name="Retrieve FASTA from NCBI" version="2.3.0">
  <description></description>
  <command><![CDATA[
    python '$__tool_directory__'/fetch_fasta_from_NCBI.py
        -i "$queryString"
        -d $dbname
        -l '$logfile'
        -c
        -o '$outfile';
    #if $dry_run == ""
        number_UIDs=\$(tail -n 2 $logfile | perl -ne '/Found (\d+) UID/ && print \$1');
        python '$__tool_directory__'/fetch_fasta_from_NCBI.py
            -i "$queryString"
            -d $dbname
            -u
            -l '$logfile'
            -o 'uid_outfile';
        UID_array=( \$(head uid_outfile) );
        array_len=\${#UID_array[@]};
        counter=0;
        number_of_groups=\$((array_len / 200000));
        modulo=\$((array_len % 200000));
        if [ "\$modulo" -gt 0 ];then
            number_of_groups=\$((number_of_groups + 1));
        fi;
        group_number=1;
        echo "----- Number of groups of batches: \$number_of_groups -----" >> $logfile;
        for ((i=0; i+200000<array_len;i+=200000)); do
            echo "----- Group number: \$group_number -----" >> $logfile;
            echo "\${UID_array[@]:\$i:99999}" > uid_list_1.txt;
            echo "\${UID_array[@]:\$((i+100000)):99999}" > uid_list_2.txt;
            python '$__tool_directory__'/fetch_fasta_from_NCBI.py
                -d $dbname
                -l '$logfile'
                -o 'tmp1_outfile'
                --UID_list uid_list_1.txt&
            python '$__tool_directory__'/fetch_fasta_from_NCBI.py
                -d $dbname
                -l 'tmp1_logfile'
                -o 'tmp2_outfile'
                --UID_list uid_list_2.txt&
            wait;
            cat tmp1_outfile tmp2_outfile>> $outfile;
            rm tmp1_outfile tmp2_outfile;
            cat tmp1_logfile >> $logfile;
            rm tmp1_logfile;
            rm uid_list_1.txt uid_list_2.txt;
            group_number=\$((group_number + 1));
            counter=\$(( counter + 200000 ));
        done;
        echo "----- Group number: \$group_number -----" >> $logfile;
        echo "----- Last group -----" >> $logfile;
        if [ "\$counter" -lt "\$array_len" ]; then
            echo "\${UID_array[@]:\$counter:\$((array_len - counter + 1))}" > uid_list.txt;
            python '$__tool_directory__'/fetch_fasta_from_NCBI.py
                -d $dbname
                -l '$logfile'
                -o 'tmp_outfile'
                --UID_list uid_list.txt;
            rm uid_list.txt;
            cat tmp_outfile >> $outfile;
            rm tmp_outfile;
        fi;
    #end if
  ]]></command>

  <inputs>
    <param name="queryString" type="text" size="5x80" area="True" value="txid10239[orgn] NOT txid131567[orgn] AND complete[all] NOT partial[title] NOT phage[title]" label="Query to NCBI in entrez format" help="exemple: Drosophila melanogaster[Organism] AND Gcn5[Title]">
      <sanitizer>
        <valid initial="string.printable">
          <remove value="&quot;"/>
          <remove value="\"/>
        </valid>
        <mapping initial="none">
          <add source="&quot;" target="\&quot;"/>
          <add source="\" target="\\"/>
        </mapping>
      </sanitizer>
    </param>
    <param name="dbname" type="select" label="NCBI database">
      <option value="nuccore">Nucleotide</option>
      <option value="protein">Protein</option>
    </param>
    <param name="dry_run" type="boolean" label="Get only the number of sequences" truevalue="--count" falsevalue="" checked="false"/>
  </inputs>
  <outputs>
    <data name="outfile" format="fasta" label="${tool.name} (${dbname.value_label}) with queryString '${queryString.value}'" >
      <filter> dry_run == False</filter>
    </data>
    <data format="txt" name="logfile" label="${tool.name}: log"/>
  </outputs>
  <tests>
      <test>
          <param name="queryString" value="9629650[gi]" />
          <param name="dbname" value="nuccore" />
          <output name="outfilename" ftype="fasta" file="output.fa" />
      </test>
      <test>
          <param name="queryString" value="CU929326[Accession]" />
          <param name="dbname" value="nuccore" />
          <param name="date_filter" value="1"/>
          <param name="dry_run" value="True"/>
          <output name="logfile" ftype="txt" file="dry_run.log" compare="sim_size"/>
      </test>
      <test>
          <param name="queryString" value="Drosophila[Organism] AND 2014[PDAT] AND virus" />
          <output name="outfilename" ftype="fasta" >
              <metadata name="sequences" value="13" />
          </output>
      </test>
  </tests>
  <help>
**What it does**

This tool retrieves nucleotide/peptide sequences from the corresponding NCBI database (nuccore or protein) for a given entrez query.

The tool is preset with "txid10239[orgn] NOT txid131567[orgn] AND complete NOT partial[title] NOT phage[title]" for metaVisitor use purpose

See `Entrez help`_ for explanation of query formats

Be sure to use the appropriate NCBI query syntax. Always use [] to specify the search fields.

By checking the checkbox you can also run your query without sequence retrieval and get the number of sequences your query will fetch.

Note that the tool may fail in case of interrupted connexion with the NCBI database (see the log dataset)

Retrieval progress is reported in the log dataset.

**Acknowledgments**

This Galaxy tool has been adapted from the galaxy tool `get_fasta_from_taxon`_.

It is Copyright © 2014-2015 `CNRS and University Pierre et Marie Curie`_ and is released under the `MIT license`_.

.. _Entrez help: https://www.ncbi.nlm.nih.gov/books/NBK3837/#EntrezHelp.Entrez_Searching_Options
.. _get_fasta_from_taxon: https://toolshed.g2.bx.psu.edu/view/crs4/get_fasta_from_taxon
.. _CNRS and University Pierre et Marie Curie: http://www.ibps.upmc.fr/en
.. _MIT license: http://opensource.org/licenses/MIT

  </help>
  <citations>
      <citation type="doi">10.1186/1471-2105-14-73</citation>
  </citations>
</tool>