annotate tools/ncbi_blast_plus/ncbi_rpstblastn_wrapper.xml @ 9:9dabbfd73c8a draft

Uploaded v0.0.19, adds wrappers for rpsblast and rpstblastn with new blastdb_d.loc file for their protein domain database. Also includes other minor improvements.

<tool id="ncbi_rpstblastn_wrapper" name="NCBI BLAST+ rpstblastn" version="0.0.3">
    <description>Search protein domain database (PSSMs) with translated nucleotide query sequence(s)</description>
    <!-- If job splitting is enabled, break up the query file into parts -->
    <parallelism method="multi" split_inputs="query" split_mode="to_size" split_size="1000" shared_inputs="subject" merge_outputs="output1"></parallelism>
    <requirements>
        <requirement type="binary">rpstblastn</requirement>
        <requirement type="package" version="2.2.26+">blast+</requirement>
    </requirements>
    <version_command>rpstblastn -version</version_command>
    <command>
## The command is a Cheetah template which allows some Python based syntax.
## Lines starting hash hash are comments. Galaxy will turn newlines into spaces
rpstblastn
-query "$query"
#if $db_opts.db_opts_selector == "db":
  -db "${db_opts.database.fields.path}"
#elif $db_opts.db_opts_selector == "histdb":
  -db "${os.path.join($db_opts.histdb.extra_files_path,'blastdb')}"
#end if
-evalue $evalue_cutoff
-out "$output1"
##Set the extended list here so if/when we add things, saved workflows are not affected
#if str($out_format)=="ext":
    -outfmt "6 std sallseqid score nident positive gaps ppos qframe sframe qseq sseq qlen slen"
#else:
    -outfmt $out_format
#end if
##Seems rpstblastn does not currently support multiple threads :(
##-num_threads 8
#if $adv_opts.adv_opts_selector=="advanced":
$adv_opts.filter_query
## Need int(str(...)) because $adv_opts.max_hits is an InputValueWrapper object not a string
## Note -max_target_seqs overrides -num_descriptions and -num_alignments
#if (str($adv_opts.max_hits) and int(str($adv_opts.max_hits)) > 0):
-max_target_seqs $adv_opts.max_hits
#end if
#if (str($adv_opts.word_size) and int(str($adv_opts.word_size)) > 0):
-word_size $adv_opts.word_size
#end if
$adv_opts.parse_deflines
## End of advanced options:
#end if
    </command>
    <stdio>
        <!-- Anything other than zero is an error -->
        <exit_code range="1:" />
        <exit_code range=":-1" />
        <!-- In case the return code has not been set propery check stderr too -->
        <regex match="Error:" />
        <regex match="Exception:" />
    </stdio>
    <inputs>
        <param name="query" type="data" format="fasta" label="Nucleotide query sequence(s)"/> 
        <conditional name="db_opts">
            <param name="db_opts_selector" type="select" label="Protein domain database (PSSM)">
              <option value="db" selected="True">Locally installed BLAST database</option>
              <!-- TODO - define new datatype
              <option value="histdb">BLAST protein domain database from your history</option>
              -->
            </param>
            <when value="db">
                <param name="database" type="select" label="Protein domain database">
                    <options from_file="blastdb_d.loc">
                      <column name="value" index="0"/>
                      <column name="name" index="1"/>
                      <column name="path" index="2"/>
                    </options>
                </param>
                <param name="histdb" type="hidden" value="" />
                <param name="subject" type="hidden" value="" />
            </when>
            <!-- TODO - define new datatype
            <when value="histdb">
                <param name="database" type="hidden" value="" />
                <param name="histdb" type="data" format="blastdbd" label="Protein domain database" />
                <param name="subject" type="hidden" value="" />
            </when>
            -->
        </conditional>
        <param name="evalue_cutoff" type="float" size="15" value="0.001" label="Set expectation value cutoff" />
        <param name="out_format" type="select" label="Output format">
            <option value="6">Tabular (standard 12 columns)</option>
            <option value="ext" selected="True">Tabular (extended 24 columns)</option>
            <option value="5">BLAST XML</option>
            <option value="0">Pairwise text</option>
            <option value="0 -html">Pairwise HTML</option>
            <option value="2">Query-anchored text</option>
            <option value="2 -html">Query-anchored HTML</option>
            <option value="4">Flat query-anchored text</option>
            <option value="4 -html">Flat query-anchored HTML</option>
            <!--
            <option value="-outfmt 11">BLAST archive format (ASN.1)</option>
            -->
        </param>
        <conditional name="adv_opts">
            <param name="adv_opts_selector" type="select" label="Advanced Options">
              <option value="basic" selected="True">Hide Advanced Options</option>
              <option value="advanced">Show Advanced Options</option>
            </param>
            <when value="basic" />
            <when value="advanced">
                <!-- Could use a select (yes, no, other) where other allows setting 'window locut hicut' -->
                <param name="filter_query" type="boolean" label="Filter out low complexity regions (with SEG)" truevalue="-seg yes" falsevalue="-seg no" checked="false" />
                <!-- Why doesn't optional override a validator? I want to accept an empty string OR a non-negative integer -->
                <param name="max_hits" type="integer" value="0" label="Maximum hits to show" help="Use zero for default limits">
                    <validator type="in_range" min="0" />
                </param>
                <!-- I'd like word_size to be optional, with minimum 2 for rpsblast -->
                <param name="word_size" type="integer" value="0" label="Word size for wordfinder algorithm" help="Use zero for default, otherwise minimum 2.">
                    <validator type="in_range" min="0" />
                </param>
                <param name="parse_deflines" type="boolean" label="Should the query and subject defline(s) be parsed?" truevalue="-parse_deflines" falsevalue="" checked="false" help="This affects the formatting of the query/subject ID strings"/>
            </when>
        </conditional>
    </inputs>
    <outputs>
        <data name="output1" format="tabular" label="rpstblastn on ${on_string}">
            <change_format>
                <when input="out_format" value="0" format="txt"/>
                <when input="out_format" value="0 -html" format="html"/>
                <when input="out_format" value="2" format="txt"/>
                <when input="out_format" value="2 -html" format="html"/>
                <when input="out_format" value="4" format="txt"/>
                <when input="out_format" value="4 -html" format="html"/>
                <when input="out_format" value="5" format="blastxml"/>
            </change_format>
        </data>
    </outputs>
    <help>
    
.. class:: warningmark

**Note**. Database searches may take a substantial amount of time.
For large input datasets it is advisable to allow overnight processing.  

-----

**What it does**

Search a *protein domain database* using a *nucleotide query*,
using the NCBI BLAST+ rpstblastn command line tool.

The protein domain databases use position-specific scoring matrices
(PSSMs) and are available for a number of domain collections including:

*CDD* - NCBI curarated meta-collection of domains, see
http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd_help.shtml#NCBI_curated_domains

*Kog* - PSSMs from automatically aligned sequences and sequence
fragments classified in the KOGs resource, the eukaryotic 
counterpart to COGs, see http://www.ncbi.nlm.nih.gov/COG/new/

*Cog* - PSSMs from automatically aligned sequences and sequence
fragments classified in the COGs resource, which focuses primarily
on prokaryotes, see http://www.ncbi.nlm.nih.gov/COG/new/

*Pfam* - PSSMs from Pfam-A seed alignment database, see
http://pfam.sanger.ac.uk/

*Smart* - PSSMs from SMART domain alignment database, see
http://smart.embl-heidelberg.de/

*Tigr* - PSSMs from TIGRFAM database of protein families, see
http://www.jcvi.org/cms/research/projects/tigrfams/overview/

*Prk* - PSSms from automatically aligned stable clusters in the
Protein Clusters database, see
http://www.ncbi.nlm.nih.gov/proteinclusters?cmd=search&amp;db=proteinclusters

The exact list of domain databases offered will depend on how your
local Galaxy has been configured.

-----

**Output format**

Because Galaxy focuses on processing tabular data, the default output of this
tool is tabular. The standard BLAST+ tabular output contains 12 columns:

====== ========= ============================================
Column NCBI name Description
------ --------- --------------------------------------------
     1 qseqid    Query Seq-id (ID of your sequence)
     2 sseqid    Subject Seq-id (ID of the database hit)
     3 pident    Percentage of identical matches
     4 length    Alignment length
     5 mismatch  Number of mismatches
     6 gapopen   Number of gap openings
     7 qstart    Start of alignment in query
     8 qend      End of alignment in query
     9 sstart    Start of alignment in subject (database hit)
    10 send      End of alignment in subject (database hit)
    11 evalue    Expectation value (E-value)
    12 bitscore  Bit score
====== ========= ============================================

The BLAST+ tools can optionally output additional columns of information,
but this takes longer to calculate. Most (but not all) of these columns are
included by selecting the extended tabular output. The extra columns are
included *after* the standard 12 columns. This is so that you can write
workflow filtering steps that accept either the 12 or 24 column tabular
BLAST output. Galaxy now uses this extended 24 column output by default.

====== ============= ===========================================
Column NCBI name     Description
------ ------------- -------------------------------------------
    13 sallseqid     All subject Seq-id(s), separated by a ';'
    14 score         Raw score
    15 nident        Number of identical matches
    16 positive      Number of positive-scoring matches
    17 gaps          Total number of gaps
    18 ppos          Percentage of positive-scoring matches
    19 qframe        Query frame
    20 sframe        Subject frame
    21 qseq          Aligned part of query sequence
    22 sseq          Aligned part of subject sequence
    23 qlen          Query sequence length
    24 slen          Subject sequence length
====== ============= ===========================================

The third option is BLAST XML output, which is designed to be parsed by
another program, and is understood by some Galaxy tools.

You can also choose several plain text or HTML output formats which are designed to be read by a person (not by another program).
The HTML versions use basic webpage formatting and can include links to the hits on the NCBI website.
The pairwise output (the default on the NCBI BLAST website) shows each match as a pairwise alignment with the query.
The two query anchored outputs show a multiple sequence alignment between the query and all the matches,
and differ in how insertions are shown (marked as insertions or with gap characters added to the other sequences).

-------

**References**

Marchler-Bauer A, Bryant SH. CD-Search: protein domain annotations on the fly. Nucleic Acids Res. 2004 Jul 1;32(Web Server issue):W327-31.

    </help>
</tool>