view ob_filter.xml @ 13:ff2dd4b7d918 draft

"planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/chemicaltoolbox/openbabel commit 944ea4bb8a9cd4244152a4a4fecd0485fabc2ad0"

<tool id="openbabel_filter" name="Filter" version="@TOOL_VERSION@+galaxy@GALAXY_VERSION@">
    <description> a set of molecules from a file</description>
    <!--parallelism method="multi" split_inputs="infile" split_mode="to_size" split_size="10000" shared_inputs="" merge_outputs="outfile"></parallelism-->
    <macros>
        <import>macros.xml</import>
        <token name="@GALAXY_VERSION@">0</token>
    </macros>
    <expand macro="requirements"/>
    <command detect_errors="aggressive">
<![CDATA[
    python '$__tool_directory__/ob_filter.py'
      -i '${infile}'
      -o '${outfile}'
      -iformat "${infile.ext}"
      -oformat "${infile.ext}"
      #if $filter_methods.filter_methods_opts == "__filter_by_name__":
          --list_of_names '$name_file'
          --filters '__filter_by_name__'
      #else
        --filters '{
          #if $filter_methods.filter_methods_opts == "ruleof5":
            "hbd" : [0, 5],
            "hba" : [0, 10],
            "molwt" : [0, 500],
            "logp" : [-5, 5],
          #elif $filter_methods.filter_methods_opts == "LeadLike":
            "rotbonds" : [0, 7],
            "molwt" : [0, 350],
            "logp" : [-5, 3.5],
          #elif $filter_methods.filter_methods_opts == "DrugLike":
            "hba" : [0, 10],
            "rotbonds" : [0, 8],
            "molwt" : [150, 500],
            "logp" : [-5, 5],
            "psa" : [0, 150],
          #elif $filter_methods.filter_methods_opts == "FragmentLike":
            "rotbonds" : [0, 5],
            "molwt" : [0, 250],
            "logp" : [-5, 2.5],
          #else:
            #for $filter in $filter_methods.filter_set:
              #if $filter.filter_sel.filter_sel_opts == "field":
                #set $filter_selected = $filter.filter_sel['field_name']
                "$filter_selected": [$filter.filter_sel['field_min'], $filter.filter_sel['field_max'] ],
              #else:
                #set $filter_selected = $filter.filter_sel.filter_sel_opts
                #set $filter_min = $filter_selected + "_min"
                #set $filter_max = $filter_selected + "_max"
                "$filter_selected" : [$filter.filter_sel[$filter_min], $filter.filter_sel[$filter_max] ],
              #end if
            #end for
          #end if
          }'
        #end if
]]>
  </command>
  <inputs>
    <param name="infile" type="data" format="sdf,smi,mol,mol2,cml,inchi"
        label="Select input file (can be previously annotated with the 'Compute physico-chemical properties' tool)"/>
    <conditional name="filter_methods">
      <param name="filter_methods_opts" type="select" label="Select a pre-defined filtering set">
        <option value="user">User-defined properties</option>
        <option value="ruleof5">Lipinski's Rule-of-Five</option>
        <option value="LeadLike">Lead-like properties</option>
        <option value="DrugLike">Drug-like properties</option>
        <option value="FragmentLike">Fragment-like properties</option>
        <option value="__filter_by_name__">Filter by molecule name</option>
      </param>
      <when value="__filter_by_name__">
            <param name="name_file" type="data" format="txt,tabular"
                label="File with molecule names that match the names in the SDF file" help="Every name one line."/>
      </when>
      <when value="ruleof5" />
      <when value="LeadLike" />
      <when value="DrugLike" />
      <when value="FragmentLike" />
      <when value="user">
        <repeat name="filter_set" title="Select custom filters">
          <conditional name="filter_sel">
            <param name="filter_sel_opts" type="select" label="Select properties to filter">
              <option value="hbd">Number of hydrogen-bond donor groups</option>
              <option value="hba">Number of hydrogen-bond acceptor groups</option>
              <option value="psa">Total polar surface area</option>
              <option value="rotbonds">Number of rotatable bonds</option>
              <option value="molwt">Molecular weight</option>
              <option value="logp">Predicted value of logP</option>
              <option value="mr">Predicted value for the molecular refractivity</option>
              <option value="atoms">Number of atoms</option>
              <option value="field">Filter by a user-defined SDF property</option>
              <!-- Add later, we need to add a new smarts pattern to plugindefines.txt
              <option value="hatoms">Number of heavy atoms</option>-->
              <!-- We skip that options, for further information please read: http://www.dalkescientific.com/writings/diary/archive/2011/06/04/dealing_with_sssr.html -->
              <!--<option value="rings">Number of rings</option>-->
            </param>
            <when value="hbd">
              <param name="hbd_min" type="integer" value="" label="Minimum number of hydrogen bond donors"/>
              <param name="hbd_max" type="integer" value="" label="Maximum number of hydrogen bond donors"/>
            </when>
            <when value="hba">
              <param name="hba_min" type="integer" value="" label="Minimum number of hydrogen bond acceptors"/>
              <param name="hba_max" type="integer" value="" label="Maximum number of hydrogen bond acceptors"/>
            </when>
            <when value="psa">
              <param name="psa_min" type="integer" value="" label="Minimum threshold for the total polar surface area"/>
              <param name="psa_max" type="integer" value="" label="Maximum threshold for the total polar surface area"/>
            </when>
            <when value="rotbonds">
              <param name="rotbonds_min" type="integer" value="" label="Minimum number of rotatable bonds"/>
              <param name="rotbonds_max" type="integer" value="" label="Maximum number of rotatable bonds"/>
            </when>
            <when value="molwt">
              <param name="molwt_min" type="integer" value="" label="Minimum threshold value for the molecular weight"/>
              <param name="molwt_max" type="integer" value="" label="Maximum threshold value for the molecular weight"/>
            </when>
            <when value="logp">
              <param name="logp_min" type="float" value="" label="Minimum threshold value for log P"/>
              <param name="logp_max" type="float" value="" label="Maximum threshold value for log P"/>
            </when>
            <when value="mr">
              <param name="mr_min" type="float" value="" label="Minimum threshold value for the molecular refractivity"/>
              <param name="mr_max" type="float" value="" label="Maximum threshold value for the molecular refractivity"/>
            </when>
            <when value="atoms">
              <param name="atoms_min" type="integer" value="" label="Minimum number of atoms"/>
              <param name="atoms_max" type="integer" value="" label="Maximum number of atoms"/>
            </when>
            <when value="field">
              <param name="field_name" type="text" value="" label="Name of the SDF property to filter by">
                <sanitizer invalid_char="">
                  <valid initial="string.ascii_letters,string.digits">
                    <add value="_" />
                    <add value="." />
                    <add value=" " />
                  </valid>
                </sanitizer>
              </param>
              <param name="field_min" type="integer" value="" label="Minimum value"/>
              <param name="field_max" type="integer" value="" label="Maximum value"/>
            </when>
            <!--<when value="rings">
              <param name="rings_min" type="integer" value="" label="Minimum number of rings"/>
              <param name="rings_max" type="integer" value="" label="Maximum number of rings"/>
            </when>-->
          </conditional>
        </repeat>
      </when>
    </conditional>
  </inputs>
  <outputs>
        <expand macro="output_like_input"/>
  </outputs>
  <tests>
    <test>
        <param name="infile" value="ligands_with_title.sdf" ftype="sdf" />
        <param name="filter_methods_opts" value="__filter_by_name__" />
        <param name="name_file" value="name_file.txt" />
        <output name="outfile" file="filterd_by_name.sdf" ftype="sdf" lines_diff="4" />
    </test>
    <test>
        <param name="infile" ftype="smi" value="CID_2244.smi"/>
        <param name="filter_methods_opts" value="ruleof5" />
        <output name="outfile" ftype="smi" file="ob_filter_on_CID2244.smi" />
    </test>
    <test>
        <param name="infile" ftype="sdf" value="CID_2244.sdf"/>
        <param name="filter_methods_opts" value="user" />
        <param name="filter_sel_opts" value="field" />
        <param name="field_name" value="PUBCHEM_EXACT_MASS"/>
        <param name="field_min" value="100"/>
        <param name="field_max" value="200"/>
        <output name="outfile" ftype="sdf" file="ob_filter_on_CID2244.sdf" lines_diff="2"/>
    </test>
    <!--
    Limitation of the test framework:
    https://trello.com/card/disambiguated-conditional-parameters-not-supported-in-unit-tests/506338ce32ae458f6d15e4b3/820
    <test>
        <param name="infile" ftype="smi" value="CID_2244.smi"/>
        <param name="filter_methods_opts" value="user" />
        <param name="filter_sel_opts" value="hba" />
        <param name="hbd_min" value="0" />
        <param name="hbd_max" value="5" />
        <output name="outfile" ftype="smi" file="ob_filter_on_CID2244_2.smi" />
    </test>
    -->
  </tests>
  <help>
<![CDATA[


.. class:: infomark

**What this tool does**

Filters a library of compounds based on user-defined physico-chemical parameters or predefined options (e.g. Ro5, lead-like properties, etc.). Multiple parameters can be selected for more specific queries.

-----

.. class:: warningmark

**Hint**

| If your input file is in SDF format you can use the *Compute physico-chemical properties* tool to precalulate the properties and use the filter on that precomputed dataset. This should be faster and the file can be reused, although it is larger than a SMILES file.
|
| For exact matches use the target value for both minimum and maximum parameters (e.g. a selection of exactly 4 rotatable bonds can be performed by selecting 4 as minimum and maximum value).
|
| Selecting the same property multiple times with different parameters will result in querying the largest overlapping subset of values for the parameter (e.g. a selection of between 0 and 3 rotatable bonds plus a selection between 2 and 4 will result in a query for compounds between 2 and 3 rotatable bonds).

-----

.. class:: infomark

**Definition of the pre-defined filtering rules**

  **# Lipinski's Rule of Five:**
    =< 5 Hydrogen-bond donor groups

    =< 10 Hydrogen-bond acceptor groups

    =< 500 Molecular weight

    =< 5 octanol/water partition coefficient (log P)

  **# Lead Like properties** (Teague, Davis, Leeson, Oprea, Angew Chem Int Ed Engl. 1999 Dec 16;38(24):3743-3748):
    =< 7 rotatable bonds

    =< 350 Molecular weight

    =< 3.5 octanol/water partition coefficient (log P)

  **# Drug Like properties** (Lipinski, J Pharmacol Toxicol Methods. 2000 Jul-Aug;44(1):235-49):
    =< 10 Hydrogen-bond acceptor groups

    =< 8 rotatable bonds

    150 =< Molecular weight =< 500

    =< 150 Polar Surface Area

    =< 5 octanol/water partition coefficient (log P)

  **# Fragment Like properties** (Carr RA, Congreve M, Murray CW, Rees DC, Drug Discov Today. 2005 Jul 15;10(14):987):
    =< 5 rotatable bonds

    =< 250 Molecular weight

    =< 2.5 octanol/water partition coefficient (log P)

-----

.. class:: infomark

**Input**

| - `SD-Format`_
| - `SMILES Format`_

.. _SD-Format: http://en.wikipedia.org/wiki/Chemical_table_file
.. _SMILES Format: http://en.wikipedia.org/wiki/Simplified_molecular_input_line_entry_specification

-----

.. class:: infomark

**Output**

| SDF formatted coordinates of the molecules, with selected properties stored as meta-data for each compound.
|
| SMILES, InChI or mol2 formatted files containing the 1D strings or 3D coordinates of each compound.



]]>
    </help>
    <expand macro="citations">
        <citation type="doi">10.1186/1752-153X-2-5</citation>
    </expand>
</tool>