annotate sort_filter.xml @ 2:a6ec6c55267e draft

"planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/rdock commit db76c3bc4ced257e2f622fcf8fcc6d2e28de3577"

<tool id="rdock_sort_filter" name="SDF sort and filter" version="0.2.0">
    <description>using the sdsort provided with rDock</description>
    <macros>
        <import>rdock_macros.xml</import>
    </macros>
    <expand macro="requirements"/>
    <command><![CDATA[

cat '$input'
#if $filter
| sdfilter -f'$filter'
#end if
#if $name_field
| sdsort -n $descending -s -f'$sort_field' -id'$name_field'
| sdfilter -f'\$_COUNT <= $top' -s'$name_field'
#end if
#if $global_sort and $sort_field
| sdsort -n $descending -f'$sort_field'
#end if
> '$output'

    ]]></command>

    <inputs>
        <param type="data" name="input" format="sdf" label="Molecules" help="Molecules in SDF format"/>

        <param name="filter" type="text" label="Filter expression" optional="true" help="Perl expression for filter">
            <sanitizer>
                <valid initial="string.printable">
                    <remove value="&apos;"/>
                    <remove value="&quot;"/>
                    <remove value="@"/>
                    <remove value="#"/>                    
                    <remove value="|"/>                    
                </valid>
                <mapping initial="none"/>
            </sanitizer>
        </param>

        <param name="sort_field" type="text" label="Field to sort on" optional="true" help="Name of the field to sort records by">
            <sanitizer>
                <valid initial="string.printable">
                    <remove value="&apos;"/>
                </valid>
                <mapping initial="none"/>
            </sanitizer>
        </param>
        <param name="descending" type="boolean" label="Sort descending" truevalue="-r" falsevalue="" checked="true"
               help="Sort ascending or descending"/>
        <param name="global_sort" type="boolean" label="Global sort" checked="true"
               help="Sort all records in file after filtering (true) or just sort within the blocks identified by $name_field (false)"/>

        <param name="name_field" type="text" label="Grouping field name" optional="true" help="Name of the field to group records by (must be sequential)">
            <sanitizer>
                <valid initial="string.printable">
                    <remove value="&apos;"/>
                </valid>
                <mapping initial="none"/>
            </sanitizer>
        </param>
        <param name="top" type="integer" value="1" label="Number of records to keep in output" optional="true" help="Number of best scoring records to keep"/>

    </inputs>
    <outputs>
        <data name="output" format="sdf" label="SDF sort+filter on ${on_string}"/>
    </outputs>
    <tests>
        <test>
            <param name="input" value="poses.sdf"/>
            <param name="sort_field" value="TransFSScore"/>
            <param name="name_field" value="Name"/>
            <param name="descending" value="True"/>
            <output name="output" file="poses-descending.sdf" ftype="sdf"/>
        </test>
        <test>
            <param name="input" value="poses.sdf"/>
            <param name="sort_field" value="TransFSScore"/>
            <param name="name_field" value="Name"/>
            <param name="descending" value="False"/>
            <output name="output" file="poses-ascending.sdf" ftype="sdf"/>
        </test>
        <test>
            <param name="input" value="poses.sdf"/>
            <param name="filter" value="$TransFSScore > 0.2"/>
            <param name="sort_field" value="TransFSScore"/>
            <param name="name_field" value="Name"/>
            <param name="descending" value="False"/>
            <output name="output" file="poses-filt-0.2.sdf" ftype="sdf"/>
        </test>
        <test>
            <param name="input" value="poses.sdf"/>
            <param name="filter" value="$TransFSScore > 0.1 and $TransFSScore > 0.2"/>
            <param name="sort_field" value="TransFSScore"/>
            <param name="name_field" value="Name"/>
            <param name="descending" value="False"/>
            <output name="output" file="poses-filt-0.2.sdf" ftype="sdf"/>
        </test>
        <test>
            <param name="input" value="poses.sdf"/>
            <param name="sort_field" value="TransFSScore"/>
            <param name="name_field" value="Name"/>
            <param name="descending" value="True"/>
            <param name="global_sort" value="False"/>
            <output name="output" file="poses-desc-noglobal.sdf" ftype="sdf"/>
        </test>
        <test>
            <param name="input" value="poses.sdf"/>
            <param name="filter" value="$TransFSScore > 0.2"/>
            <output name="output" file="poses-filt-only.sdf" ftype="sdf"/>
        </test>
    </tests>
    <help><![CDATA[

.. class:: infomark

This tool sorts and filters SD files using the sdsort and sdfilter tools from the rDock suite.
See http://rdock.sourceforge.net/ for more details about rDock and associated programs.
The expected use is for filtering and sorting virtual screening results such as docking.

-----

.. class:: infomark

**Inputs**
An SD-file, together with names of fields to filter, sort and group records by, and the number of records to appear in the output.

An optional filter can be specified that is first applied to the records. This filter (the 'filter' parameter) must be
specified as required by the 'sdfilter' application (see http://rdock.sourceforge.net/wp-content/uploads/2015/08/rDock_User_Guide.pdf)
which is a Perl expression. As an example, if your SDF has a field name 'SCORE' which has numeric values then a valid
filter expression would be '$SCORE > 90' (note the $ symbol).
If you require to use multiple filters then you can combine them in a single expression like this:
'$A < 5 and $B <7', or '$A < 5 or $B <7'

The sorting is then performed on groups of molecules, with the groups being identified by a field in the SD-file (the 'name_field'
parameter). Records from a group MUST be sequential in the input file. If 'name_field' is not specified then this grouping
and sorting step is skipped. Sorting is performed by the rDock 'sdsort' application.
The records within each group are sorted by the value of a field in the SD-file (the 'sort_field' parameter) and you can
specify ascending or descending order (the 'descending' parameter).
Then a number of top scoring (the 'top' parameter, typically having a value of 1) are retained.

Finally, if the 'global_sort' parameter is set to 'True' then the all the records remaining are sorted according to the
'sort_field' and 'descending' parameters. Note: this step can use lots of memory if the files are very big.

-----

.. class:: infomark

**Outputs**
An SD-file, containing molecules filtered and sorted according to the parameters.

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