Mercurial > repos > peterjc > ncbi_blast_plus

comparison ncbi_blast_plus/ncbi_blastp_wrapper.xml @ 20:688f3fb09a6a draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded v0.0.20 preview 11, moved to GitHub, MIT license, reST markup.
author peterjc
date Tue, 30 Jul 2013 07:33:46 -0400
parents
children 61f402b6e240
comparison
equal deleted inserted replaced
19:c1a6e5aefee0 20:688f3fb09a6a
1 <tool id="ncbi_blastp_wrapper" name="NCBI BLAST+ blastp" version="0.0.20">
2 <description>Search protein database with protein query sequence(s)</description>
3 <!-- If job splitting is enabled, break up the query file into parts -->
4 <parallelism method="multi" split_inputs="query" split_mode="to_size" split_size="1000" shared_inputs="subject,histdb" merge_outputs="output1"></parallelism>
5 <requirements>
6 <requirement type="binary">blastp</requirement>
7 <requirement type="package" version="2.2.26+">blast+</requirement>
8 </requirements>
9 <version_command>blastp -version</version_command>
10 <command>
11 ## The command is a Cheetah template which allows some Python based syntax.
12 ## Lines starting hash hash are comments. Galaxy will turn newlines into spaces
13 blastp
14 -query "$query"
15 #if $db_opts.db_opts_selector == "db":
16 -db "${db_opts.database.fields.path}"
17 #elif $db_opts.db_opts_selector == "histdb":
18 -db "${os.path.join($db_opts.histdb.extra_files_path,'blastdb')}"
19 #else:
20 -subject "$db_opts.subject"
21 #end if
22 -task $blast_type
23 -evalue $evalue_cutoff
24 -out "$output1"
25 ##Set the extended list here so if/when we add things, saved workflows are not affected
26 #if str($out_format)=="ext":
27 -outfmt "6 std sallseqid score nident positive gaps ppos qframe sframe qseq sseq qlen slen"
28 #else:
29 -outfmt $out_format
30 #end if
31 -num_threads 8
32 #if $adv_opts.adv_opts_selector=="advanced":
33 $adv_opts.filter_query
34 -matrix $adv_opts.matrix
35 ## Need int(str(...)) because $adv_opts.max_hits is an InputValueWrapper object not a string
36 ## Note -max_target_seqs overrides -num_descriptions and -num_alignments
37 #if (str($adv_opts.max_hits) and int(str($adv_opts.max_hits)) > 0):
38 -max_target_seqs $adv_opts.max_hits
39 #end if
40 #if (str($adv_opts.word_size) and int(str($adv_opts.word_size)) > 0):
41 -word_size $adv_opts.word_size
42 #end if
43 ##Ungapped disabled for now - see comments below
44 ##$adv_opts.ungapped
45 $adv_opts.parse_deflines
46 ## End of advanced options:
47 #end if
48 </command>
49 <stdio>
50 <!-- Anything other than zero is an error -->
51 <exit_code range="1:" />
52 <exit_code range=":-1" />
53 <!-- In case the return code has not been set propery check stderr too -->
54 <regex match="Error:" />
55 <regex match="Exception:" />
56 </stdio>
57 <inputs>
58 <param name="query" type="data" format="fasta" label="Protein query sequence(s)"/>
59 <conditional name="db_opts">
60 <param name="db_opts_selector" type="select" label="Subject database/sequences">
61 <option value="db" selected="True">Locally installed BLAST database</option>
62 <option value="histdb">BLAST database from your history</option>
63 <option value="file">FASTA file from your history (see warning note below)</option>
64 </param>
65 <when value="db">
66 <param name="database" type="select" label="Protein BLAST database">
67 <options from_file="blastdb_p.loc">
68 <column name="value" index="0"/>
69 <column name="name" index="1"/>
70 <column name="path" index="2"/>
71 </options>
72 </param>
73 <param name="histdb" type="hidden" value="" />
74 <param name="subject" type="hidden" value="" />
75 </when>
76 <when value="histdb">
77 <param name="database" type="hidden" value="" />
78 <param name="histdb" type="data" format="blastdbp" label="Protein BLAST database" />
79 <param name="subject" type="hidden" value="" />
80 </when>
81 <when value="file">
82 <param name="database" type="hidden" value="" />
83 <param name="histdb" type="hidden" value="" />
84 <param name="subject" type="data" format="fasta" label="Protein FASTA file to use as database"/>
85 </when>
86 </conditional>
87 <param name="blast_type" type="select" display="radio" label="Type of BLAST">
88 <option value="blastp">blastp</option>
89 <option value="blastp-short">blastp-short</option>
90 </param>
91 <param name="evalue_cutoff" type="float" size="15" value="0.001" label="Set expectation value cutoff" />
92 <param name="out_format" type="select" label="Output format">
93 <option value="6">Tabular (standard 12 columns)</option>
94 <option value="ext" selected="True">Tabular (extended 24 columns)</option>
95 <option value="5">BLAST XML</option>
96 <option value="0">Pairwise text</option>
97 <option value="0 -html">Pairwise HTML</option>
98 <option value="2">Query-anchored text</option>
99 <option value="2 -html">Query-anchored HTML</option>
100 <option value="4">Flat query-anchored text</option>
101 <option value="4 -html">Flat query-anchored HTML</option>
102 <!--
103 <option value="-outfmt 11">BLAST archive format (ASN.1)</option>
104 -->
105 </param>
106 <conditional name="adv_opts">
107 <param name="adv_opts_selector" type="select" label="Advanced Options">
108 <option value="basic" selected="True">Hide Advanced Options</option>
109 <option value="advanced">Show Advanced Options</option>
110 </param>
111 <when value="basic" />
112 <when value="advanced">
113 <!-- Could use a select (yes, no, other) where other allows setting 'window locut hicut' -->
114 <param name="filter_query" type="boolean" label="Filter out low complexity regions (with SEG)" truevalue="-seg yes" falsevalue="-seg no" checked="false" />
115 <param name="matrix" type="select" label="Scoring matrix">
116 <option value="BLOSUM90">BLOSUM90</option>
117 <option value="BLOSUM80">BLOSUM80</option>
118 <option value="BLOSUM62" selected="true">BLOSUM62 (default)</option>
119 <option value="BLOSUM50">BLOSUM50</option>
120 <option value="BLOSUM45">BLOSUM45</option>
121 <option value="PAM250">PAM250</option>
122 <option value="PAM70">PAM70</option>
123 <option value="PAM30">PAM30</option>
124 </param>
125 <!-- Why doesn't optional override a validator? I want to accept an empty string OR a non-negative integer -->
126 <param name="max_hits" type="integer" value="0" label="Maximum hits to show" help="Use zero for default limits">
127 <validator type="in_range" min="0" />
128 </param>
129 <!-- I'd like word_size to be optional, with minimum 2 for blastp -->
130 <param name="word_size" type="integer" value="0" label="Word size for wordfinder algorithm" help="Use zero for default, otherwise minimum 2.">
131 <validator type="in_range" min="0" />
132 </param>
133 <!--
134 Can't use '-ungapped' on its own, error back is:
135 Composition-adjusted searched are not supported with an ungapped search, please add -comp_based_stats F or do a gapped search
136 Tried using '-ungapped -comp_based_stats F' and blastp crashed with 'Attempt to access NULL pointer.'
137 <param name="ungapped" type="boolean" label="Perform ungapped alignment only?" truevalue="-ungapped -comp_based_stats F" falsevalue="" checked="false" />
138 -->
139 <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"/>
140 </when>
141 </conditional>
142 </inputs>
143 <outputs>
144 <data name="output1" format="tabular" label="${blast_type.value_label} on ${on_string}">
145 <change_format>
146 <when input="out_format" value="0" format="txt"/>
147 <when input="out_format" value="0 -html" format="html"/>
148 <when input="out_format" value="2" format="txt"/>
149 <when input="out_format" value="2 -html" format="html"/>
150 <when input="out_format" value="4" format="txt"/>
151 <when input="out_format" value="4 -html" format="html"/>
152 <when input="out_format" value="5" format="blastxml"/>
153 </change_format>
154 </data>
155 </outputs>
156 <tests>
157 <test>
158 <param name="query" value="four_human_proteins.fasta" ftype="fasta" />
159 <param name="db_opts_selector" value="file" />
160 <param name="subject" value="rhodopsin_proteins.fasta" ftype="fasta" />
161 <param name="database" value="" />
162 <param name="evalue_cutoff" value="1e-8" />
163 <param name="blast_type" value="blastp" />
164 <param name="out_format" value="5" />
165 <param name="adv_opts_selector" value="advanced" />
166 <param name="filter_query" value="False" />
167 <param name="matrix" value="BLOSUM62" />
168 <param name="max_hits" value="0" />
169 <param name="word_size" value="0" />
170 <param name="parse_deflines" value="True" />
171 <output name="output1" file="blastp_four_human_vs_rhodopsin.xml" ftype="blastxml" />
172 </test>
173 <test>
174 <param name="query" value="four_human_proteins.fasta" ftype="fasta" />
175 <param name="db_opts_selector" value="file" />
176 <param name="subject" value="rhodopsin_proteins.fasta" ftype="fasta" />
177 <param name="database" value="" />
178 <param name="evalue_cutoff" value="1e-8" />
179 <param name="blast_type" value="blastp" />
180 <param name="out_format" value="6" />
181 <param name="adv_opts_selector" value="advanced" />
182 <param name="filter_query" value="False" />
183 <param name="matrix" value="BLOSUM62" />
184 <param name="max_hits" value="0" />
185 <param name="word_size" value="0" />
186 <param name="parse_deflines" value="True" />
187 <output name="output1" file="blastp_four_human_vs_rhodopsin.tabular" ftype="tabular" />
188 </test>
189 <test>
190 <param name="query" value="four_human_proteins.fasta" ftype="fasta" />
191 <param name="db_opts_selector" value="file" />
192 <param name="subject" value="rhodopsin_proteins.fasta" ftype="fasta" />
193 <param name="database" value="" />
194 <param name="evalue_cutoff" value="1e-8" />
195 <param name="blast_type" value="blastp" />
196 <param name="out_format" value="ext" />
197 <param name="adv_opts_selector" value="advanced" />
198 <param name="filter_query" value="False" />
199 <param name="matrix" value="BLOSUM62" />
200 <param name="max_hits" value="0" />
201 <param name="word_size" value="0" />
202 <param name="parse_deflines" value="True" />
203 <output name="output1" file="blastp_four_human_vs_rhodopsin_ext.tabular" ftype="tabular" />
204 </test>
205 <test>
206 <param name="query" value="rhodopsin_proteins.fasta" ftype="fasta" />
207 <param name="db_opts_selector" value="file" />
208 <param name="subject" value="four_human_proteins.fasta" ftype="fasta" />
209 <param name="database" value="" />
210 <param name="evalue_cutoff" value="1e-8" />
211 <param name="blast_type" value="blastp" />
212 <param name="out_format" value="6" />
213 <param name="adv_opts_selector" value="basic" />
214 <output name="output1" file="blastp_rhodopsin_vs_four_human.tabular" ftype="tabular" />
215 </test>
216 </tests>
217 <help>
218
219 .. class:: warningmark
220
221 **Note**. Database searches may take a substantial amount of time.
222 For large input datasets it is advisable to allow overnight processing.
223
224 -----
225
226 **What it does**
227
228 Search a *protein database* using a *protein query*,
229 using the NCBI BLAST+ blastp command line tool.
230
231 .. class:: warningmark
232
233 You can also search against a FASTA file of subject protein
234 sequences. This is *not* advised because it is slower (only one
235 CPU is used), but more importantly gives e-values for pairwise
236 searches (very small e-values which will look overly signficiant).
237 In most cases you should instead turn the other FASTA file into a
238 database first using *makeblastdb* and search against that.
239
240 -----
241
242 **Output format**
243
244 Because Galaxy focuses on processing tabular data, the default output of this
245 tool is tabular. The standard BLAST+ tabular output contains 12 columns:
246
247 ====== ========= ============================================
248 Column NCBI name Description
249 ------ --------- --------------------------------------------
250 1 qseqid Query Seq-id (ID of your sequence)
251 2 sseqid Subject Seq-id (ID of the database hit)
252 3 pident Percentage of identical matches
253 4 length Alignment length
254 5 mismatch Number of mismatches
255 6 gapopen Number of gap openings
256 7 qstart Start of alignment in query
257 8 qend End of alignment in query
258 9 sstart Start of alignment in subject (database hit)
259 10 send End of alignment in subject (database hit)
260 11 evalue Expectation value (E-value)
261 12 bitscore Bit score
262 ====== ========= ============================================
263
264 The BLAST+ tools can optionally output additional columns of information,
265 but this takes longer to calculate. Most (but not all) of these columns are
266 included by selecting the extended tabular output. The extra columns are
267 included *after* the standard 12 columns. This is so that you can write
268 workflow filtering steps that accept either the 12 or 24 column tabular
269 BLAST output. Galaxy now uses this extended 24 column output by default.
270
271 ====== ============= ===========================================
272 Column NCBI name Description
273 ------ ------------- -------------------------------------------
274 13 sallseqid All subject Seq-id(s), separated by a ';'
275 14 score Raw score
276 15 nident Number of identical matches
277 16 positive Number of positive-scoring matches
278 17 gaps Total number of gaps
279 18 ppos Percentage of positive-scoring matches
280 19 qframe Query frame
281 20 sframe Subject frame
282 21 qseq Aligned part of query sequence
283 22 sseq Aligned part of subject sequence
284 23 qlen Query sequence length
285 24 slen Subject sequence length
286 ====== ============= ===========================================
287
288 The third option is BLAST XML output, which is designed to be parsed by
289 another program, and is understood by some Galaxy tools.
290
291 You can also choose several plain text or HTML output formats which are designed to be read by a person (not by another program).
292 The HTML versions use basic webpage formatting and can include links to the hits on the NCBI website.
293 The pairwise output (the default on the NCBI BLAST website) shows each match as a pairwise alignment with the query.
294 The two query anchored outputs show a multiple sequence alignment between the query and all the matches,
295 and differ in how insertions are shown (marked as insertions or with gap characters added to the other sequences).
296
297 -------
298
299 **References**
300
301 Altschul et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. 1997. Nucleic Acids Res. 25:3389-3402.
302
303 Schaffer et al. Improving the accuracy of PSI-BLAST protein database searches with composition-based statistics and other refinements. 2001. Nucleic Acids Res. 29:2994-3005.
304
305 This wrapper is available to install into other Galaxy Instances via the Galaxy
306 Tool Shed at http://toolshed.g2.bx.psu.edu/view/devteam/ncbi_blast_plus
307 </help>
308 </tool>