/tools/ncbi_blast_plus/ncbi_blastp_wrapper.xml

https://bitbucket.org/cistrome/cistrome-harvard/ · XML · 276 lines · 246 code · 18 blank · 12 comment · 0 complexity · 4b7985cccb23bb316af8a52ac3cd1bef MD5 · raw file 

    

  1. <tool id="ncbi_blastp_wrapper" name="NCBI BLAST+ blastp" version="0.0.11">
    2. 

  2.     <description>Search protein database with protein query sequence(s)</description>
    3. 

  3.     <version_command>blastp -version</version_command>
    4. 

  4.     <command interpreter="python">hide_stderr.py
    5. 

  5. ## The command is a Cheetah template which allows some Python based syntax.
    6. 

  6. ## Lines starting hash hash are comments. Galaxy will turn newlines into spaces
    7. 

  7. blastp
    8. 

  8. -query "$query"
    9. 

  9. #if $db_opts.db_opts_selector == "db":
    10. 

  10.   -db "${db_opts.database.fields.path}"
    11. 

  11. #else:
    12. 

  12.   -subject "$db_opts.subject"
    13. 

  13. #end if
    14. 

  14. -task $blast_type
    15. 

  15. -evalue $evalue_cutoff
    16. 

  16. -out $output1
    17. 

  17. ##Set the extended list here so if/when we add things, saved workflows are not affected
    18. 

  18. #if str($out_format)=="ext":
    19. 

  19.     -outfmt "6 std sallseqid score nident positive gaps ppos qframe sframe qseq sseq qlen slen"
    20. 

  20. #else:
    21. 

  21.     -outfmt $out_format
    22. 

  22. #end if
    23. 

  23. -num_threads 8
    24. 

  24. #if $adv_opts.adv_opts_selector=="advanced":
    25. 

  25. $adv_opts.filter_query
    26. 

  26. -matrix $adv_opts.matrix
    27. 

  27. ## Need int(str(...)) because $adv_opts.max_hits is an InputValueWrapper object not a string
    28. 

  28. ## Note -max_target_seqs overrides -num_descriptions and -num_alignments
    29. 

  29. #if (str($adv_opts.max_hits) and int(str($adv_opts.max_hits)) > 0):
    30. 

  30. -max_target_seqs $adv_opts.max_hits
    31. 

  31. #end if
    32. 

  32. #if (str($adv_opts.word_size) and int(str($adv_opts.word_size)) > 0):
    33. 

  33. -word_size $adv_opts.word_size
    34. 

  34. #end if
    35. 

  35. ##Ungapped disabled for now - see comments below
    36. 

  36. ##$adv_opts.ungapped
    37. 

  37. $adv_opts.parse_deflines
    38. 

  38. ## End of advanced options:
    39. 

  39. #end if
    40. 

  40.     </command>
    41. 

  41.     <inputs>
    42. 

  42.         <param name="query" type="data" format="fasta" label="Protein query sequence(s)"/> 
    43. 

  43.         <conditional name="db_opts">
    44. 

  44.             <param name="db_opts_selector" type="select" label="Subject database/sequences">
    45. 

  45.               <option value="db" selected="True">BLAST Database</option>
    46. 

  46.               <option value="file">FASTA file</option>
    47. 

  47.             </param>
    48. 

  48.             <when value="db">
    49. 

  49.                 <param name="database" type="select" label="Protein BLAST database">
    50. 

  50.                     <options from_file="blastdb_p.loc">
    51. 

  51.                       <column name="value" index="0"/>
    52. 

  52.                       <column name="name" index="1"/>
    53. 

  53.                       <column name="path" index="2"/>
    54. 

  54.                     </options>
    55. 

  55.                 </param>
    56. 

  56.                 <param name="subject" type="hidden" value="" /> 
    57. 

  57.             </when>
    58. 

  58.             <when value="file">
    59. 

  59.                 <param name="database" type="hidden" value="" /> 
    60. 

  60.                 <param name="subject" type="data" format="fasta" label="Protein FASTA file to use as database"/> 
    61. 

  61.             </when>
    62. 

  62.         </conditional>
    63. 

  63.         <param name="blast_type" type="select" display="radio" label="Type of BLAST">
    64. 

  64.             <option value="blastp">blastp</option>
    65. 

  65.             <option value="blastp-short">blastp-short</option>
    66. 

  66.         </param>
    67. 

  67.         <param name="evalue_cutoff" type="float" size="15" value="0.001" label="Set expectation value cutoff" />
    68. 

  68.         <param name="out_format" type="select" label="Output format">
    69. 

  69.             <option value="6" selected="True">Tabular (standard 12 columns)</option>
    70. 

  70.             <option value="ext">Tabular (extended 24 columns)</option>
    71. 

  71.             <option value="5">BLAST XML</option>
    72. 

  72.             <option value="0">Pairwise text</option>
    73. 

  73.             <option value="0 -html">Pairwise HTML</option>
    74. 

  74.             <option value="2">Query-anchored text</option>
    75. 

  75.             <option value="2 -html">Query-anchored HTML</option>
    76. 

  76.             <option value="4">Flat query-anchored text</option>
    77. 

  77.             <option value="4 -html">Flat query-anchored HTML</option>
    78. 

  78.             <!--
    79. 

  79.             <option value="-outfmt 11">BLAST archive format (ASN.1)</option>
    80. 

  80.             -->
    81. 

  81.         </param>
    82. 

  82.         <conditional name="adv_opts">
    83. 

  83.             <param name="adv_opts_selector" type="select" label="Advanced Options">
    84. 

  84.               <option value="basic" selected="True">Hide Advanced Options</option>
    85. 

  85.               <option value="advanced">Show Advanced Options</option>
    86. 

  86.             </param>
    87. 

  87.             <when value="basic" />
    88. 

  88.             <when value="advanced">
    89. 

  89.                 <!-- Could use a select (yes, no, other) where other allows setting 'window locut hicut' -->
    90. 

  90.                 <param name="filter_query" type="boolean" label="Filter out low complexity regions (with SEG)" truevalue="-seg yes" falsevalue="-seg no" checked="false" />
    91. 

  91.                 <param name="matrix" type="select" label="Scoring matrix">
    92. 

  92.                     <option value="BLOSUM90">BLOSUM90</option>
    93. 

  93.                     <option value="BLOSUM80">BLOSUM80</option>
    94. 

  94.                     <option value="BLOSUM62" selected="true">BLOSUM62 (default)</option>
    95. 

  95.                     <option value="BLOSUM50">BLOSUM50</option> 
    96. 

  96.                     <option value="BLOSUM45">BLOSUM45</option>
    97. 

  97.                     <option value="PAM250">PAM250</option>
    98. 

  98.                     <option value="PAM70">PAM70</option>
    99. 

  99.                     <option value="PAM30">PAM30</option>
    100. 

  100.                 </param>
    101. 

  101.                 <!-- Why doesn't optional override a validator? I want to accept an empty string OR a non-negative integer -->
    102. 

  102.                 <param name="max_hits" type="integer" value="0" label="Maximum hits to show" help="Use zero for default limits">
    103. 

  103.                     <validator type="in_range" min="0" />
    104. 

  104.                 </param>
    105. 

  105.                 <!-- I'd like word_size to be optional, with minimum 2 for blastp -->
    106. 

  106.                 <param name="word_size" type="integer" value="0" label="Word size for wordfinder algorithm" help="Use zero for default, otherwise minimum 2.">
    107. 

  107.                     <validator type="in_range" min="0" />
    108. 

  108.                 </param>
    109. 

  109.                 <!--
    110. 

  110.                 Can't use '-ungapped' on its own, error back is:
    111. 

  111.                 Composition-adjusted searched are not supported with an ungapped search, please add -comp_based_stats F or do a gapped search
    112. 

  112.                 Tried using '-ungapped -comp_based_stats F' and blastp crashed with 'Attempt to access NULL pointer.'
    113. 

  113.                 <param name="ungapped" type="boolean" label="Perform ungapped alignment only?" truevalue="-ungapped -comp_based_stats F" falsevalue="" checked="false" />
    114. 

  114.                 -->
    115. 

  115.                 <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"/>
    116. 

  116.             </when>
    117. 

  117.         </conditional>
    118. 

  118.     </inputs>
    119. 

  119.     <outputs>
    120. 

  120.         <data name="output1" format="tabular" label="${blast_type.value_label} on ${db_opts.db_opts_selector}">
    121. 

  121.             <change_format>
    122. 

  122.                 <when input="out_format" value="0" format="txt"/>
    123. 

  123.                 <when input="out_format" value="0 -html" format="html"/>
    124. 

  124.                 <when input="out_format" value="2" format="txt"/>
    125. 

  125.                 <when input="out_format" value="2 -html" format="html"/>
    126. 

  126.                 <when input="out_format" value="4" format="txt"/>
    127. 

  127.                 <when input="out_format" value="4 -html" format="html"/>
    128. 

  128.                 <when input="out_format" value="5" format="blastxml"/>
    129. 

  129.             </change_format>
    130. 

  130.         </data>
    131. 

  131.     </outputs>
    132. 

  132.     <requirements>
    133. 

  133.         <requirement type="binary">blastp</requirement>
    134. 

  134.     </requirements>
    135. 

  135.     <tests>
    136. 

  136.         <test>
    137. 

  137.             <param name="query" value="four_human_proteins.fasta" ftype="fasta" />
    138. 

  138.             <param name="db_opts_selector" value="file" />
    139. 

  139.             <param name="subject" value="rhodopsin_proteins.fasta" ftype="fasta" />
    140. 

  140.             <param name="database" value="" />
    141. 

  141.             <param name="evalue_cutoff" value="1e-8" />
    142. 

  142.             <param name="blast_type" value="blastp" />
    143. 

  143.             <param name="out_format" value="5" />
    144. 

  144.             <param name="adv_opts_selector" value="advanced" />
    145. 

  145.             <param name="filter_query" value="False" />
    146. 

  146.             <param name="matrix" value="BLOSUM62" />
    147. 

  147.             <param name="max_hits" value="0" />
    148. 

  148.             <param name="word_size" value="0" />
    149. 

  149.             <param name="parse_deflines" value="True" />
    150. 

  150.             <output name="output1" file="blastp_four_human_vs_rhodopsin.xml" ftype="blastxml" />
    151. 

  151.         </test>
    152. 

  152.         <test>
    153. 

  153.             <param name="query" value="four_human_proteins.fasta" ftype="fasta" />
    154. 

  154.             <param name="db_opts_selector" value="file" />
    155. 

  155.             <param name="subject" value="rhodopsin_proteins.fasta" ftype="fasta" />
    156. 

  156.             <param name="database" value="" />
    157. 

  157.             <param name="evalue_cutoff" value="1e-8" />
    158. 

  158.             <param name="blast_type" value="blastp" />
    159. 

  159.             <param name="out_format" value="6" />
    160. 

  160.             <param name="adv_opts_selector" value="advanced" />
    161. 

  161.             <param name="filter_query" value="False" />
    162. 

  162.             <param name="matrix" value="BLOSUM62" />
    163. 

  163.             <param name="max_hits" value="0" />
    164. 

  164.             <param name="word_size" value="0" />
    165. 

  165.             <param name="parse_deflines" value="True" />
    166. 

  166.             <output name="output1" file="blastp_four_human_vs_rhodopsin.tabular" ftype="tabular" />
    167. 

  167.         </test>
    168. 

  168.         <test>
    169. 

  169.             <param name="query" value="four_human_proteins.fasta" ftype="fasta" />
    170. 

  170.             <param name="db_opts_selector" value="file" />
    171. 

  171.             <param name="subject" value="rhodopsin_proteins.fasta" ftype="fasta" />
    172. 

  172.             <param name="database" value="" />
    173. 

  173.             <param name="evalue_cutoff" value="1e-8" />
    174. 

  174.             <param name="blast_type" value="blastp" />
    175. 

  175.             <param name="out_format" value="ext" />
    176. 

  176.             <param name="adv_opts_selector" value="advanced" />
    177. 

  177.             <param name="filter_query" value="False" />
    178. 

  178.             <param name="matrix" value="BLOSUM62" />
    179. 

  179.             <param name="max_hits" value="0" />
    180. 

  180.             <param name="word_size" value="0" />
    181. 

  181.             <param name="parse_deflines" value="True" />
    182. 

  182.             <output name="output1" file="blastp_four_human_vs_rhodopsin_ext.tabular" ftype="tabular" />
    183. 

  183.         </test>
    184. 

  184.         <test>
    185. 

  185.             <param name="query" value="rhodopsin_proteins.fasta" ftype="fasta" />
    186. 

  186.             <param name="db_opts_selector" value="file" />
    187. 

  187.             <param name="subject" value="four_human_proteins.fasta" ftype="fasta" />
    188. 

  188.             <param name="database" value="" />
    189. 

  189.             <param name="evalue_cutoff" value="1e-8" />
    190. 

  190.             <param name="blast_type" value="blastp" />
    191. 

  191.             <param name="out_format" value="6" />
    192. 

  192.             <param name="adv_opts_selector" value="basic" />
    193. 

  193.             <output name="output1" file="blastp_rhodopsin_vs_four_human.tabular" ftype="tabular" />
    194. 

  194.         </test>
    195. 

  195.     </tests>
    196. 

  196.     <help>
    197. 

  197.     
    198. 

  198. .. class:: warningmark
    199. 

  199. 

  200. **Note**. Database searches may take a substantial amount of time.
    201. 

  201. For large input datasets it is advisable to allow overnight processing.  
    202. 

  202. 

  203. -----
    204. 

  204. 

  205. **What it does**
    206. 

  206. 

  207. Search a *protein database* using a *protein query*,
    208. 

  208. using the NCBI BLAST+ blastp command line tool.
    209. 

  209. 

  210. -----
    211. 

  211. 

  212. **Output format**
    213. 

  213. 

  214. Because Galaxy focuses on processing tabular data, the default output of this
    215. 

  215. tool is tabular. The standard BLAST+ tabular output contains 12 columns:
    216. 

  216. 

  217. ====== ========= ============================================
    218. 

  218. Column NCBI name Description
    219. 

  219. ------ --------- --------------------------------------------
    220. 

  220.      1 qseqid    Query Seq-id (ID of your sequence)
    221. 

  221.      2 sseqid    Subject Seq-id (ID of the database hit)
    222. 

  222.      3 pident    Percentage of identical matches
    223. 

  223.      4 length    Alignment length
    224. 

  224.      5 mismatch  Number of mismatches
    225. 

  225.      6 gapopen   Number of gap openings
    226. 

  226.      7 qstart    Start of alignment in query
    227. 

  227.      8 qend      End of alignment in query
    228. 

  228.      9 sstart    Start of alignment in subject (database hit)
    229. 

  229.     10 send      End of alignment in subject (database hit)
    230. 

  230.     11 evalue    Expectation value (E-value)
    231. 

  231.     12 bitscore  Bit score
    232. 

  232. ====== ========= ============================================
    233. 

  233. 

  234. The BLAST+ tools can optionally output additional columns of information,
    235. 

  235. but this takes longer to calculate. Most (but not all) of these columns are
    236. 

  236. included by selecting the extended tabular output. The extra columns are
    237. 

  237. included *after* the standard 12 columns. This is so that you can write
    238. 

  238. workflow filtering steps that accept either the 12 or 24 column tabular
    239. 

  239. BLAST output.
    240. 

  240. 

  241. ====== ============= ===========================================
    242. 

  242. Column NCBI name     Description
    243. 

  243. ------ ------------- -------------------------------------------
    244. 

  244.     13 sallseqid     All subject Seq-id(s), separated by a ';'
    245. 

  245.     14 score         Raw score
    246. 

  246.     15 nident        Number of identical matches
    247. 

  247.     16 positive      Number of positive-scoring matches
    248. 

  248.     17 gaps          Total number of gaps
    249. 

  249.     18 ppos          Percentage of positive-scoring matches
    250. 

  250.     19 qframe        Query frame
    251. 

  251.     20 sframe        Subject frame
    252. 

  252.     21 qseq          Aligned part of query sequence
    253. 

  253.     22 sseq          Aligned part of subject sequence
    254. 

  254.     23 qlen          Query sequence length
    255. 

  255.     24 slen          Subject sequence length
    256. 

  256. ====== ============= ===========================================
    257. 

  257. 

  258. The third option is BLAST XML output, which is designed to be parsed by
    259. 

  259. another program, and is understood by some Galaxy tools.
    260. 

  260. 

  261. You can also choose several plain text or HTML output formats which are designed to be read by a person (not by another program).
    262. 

  262. The HTML versions use basic webpage formatting and can include links to the hits on the NCBI website.
    263. 

  263. The pairwise output (the default on the NCBI BLAST website) shows each match as a pairwise alignment with the query.
    264. 

  264. The two query anchored outputs show a multiple sequence alignment between the query and all the matches,
    265. 

  265. and differ in how insertions are shown (marked as insertions or with gap characters added to the other sequences).
    266. 

  266. 

  267. -------
    268. 

  268. 

  269. **References**
    270. 

  270. 

  271. Altschul et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. 1997. Nucleic Acids Res. 25:3389-3402.
    272. 

  272. 

  273. 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.
    274. 

  274. 

  275.     </help>
    276. 

  276. </tool>
    277.