Mercurial > repos > oinizan > demultiplex

comparison demultiplex.py.save2 @ 0:da4101033e10 draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload
author oinizan
date Wed, 18 Oct 2017 05:30:40 -0400
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:da4101033e10
1 #!/usr/bin/env python2.7
2 #
3 # Copyright (C) 2014 INRA
4 #
5 # This program is free software: you can redistribute it and/or modify
6 # it under the terms of the GNU General Public License as published by
7 # the Free Software Foundation, either version 3 of the License, or
8 # (at your option) any later version.
9 #
10 # This program is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 # GNU General Public License for more details.
14 #
15 # You should have received a copy of the GNU General Public License
16 # along with this program. If not, see <http://www.gnu.org/licenses/>.
17 #
18
19 __author__ = 'Plateforme bioinformatique Toulouse and SIGENAE'
20 __copyright__ = 'Copyright (C) 2015 INRA'
21 __license__ = 'GNU General Public License'
22 __version__ = '1.2.0'
23 __email__ = 'frogs@inra.fr'
24 __status__ = 'prod'
25
26 import os
27 import sys
28 import gzip
29 import time
30 import tarfile
31 import argparse
32
33 CURRENT_DIR = os.path.dirname(os.path.abspath(__file__))
34 # PATH
35 BIN_DIR = os.path.abspath(os.path.join(os.path.dirname(CURRENT_DIR), "libexec"))
36 os.environ['PATH'] = BIN_DIR + os.pathsep + os.environ['PATH']
37 # PYTHONPATH
38 LIB_DIR = os.path.abspath(os.path.join(os.path.dirname(CURRENT_DIR), "lib"))
39 sys.path.append(LIB_DIR)
40 if os.getenv('PYTHONPATH') is None: os.environ['PYTHONPATH'] = LIB_DIR
41 else: os.environ['PYTHONPATH'] = os.environ['PYTHONPATH'] + os.pathsep + LIB_DIR
42
43 from frogsUtils import *
44
45
46 ##################################################################################################################################################
47 #
48 # COMMAND LINES
49 #
50 ##################################################################################################################################################
51 class Demultiplex(Cmd):
52 """
53 @summary : Demultiplex samples.
54 """
55 def __init__(self, R1_input_file, R2_input_file, barcode_file, mismatches, end, global_tmp_files, R1_output_files, R2_output_files, demultiplex_err_files1, demultiplex_err_files2, demultiplex_log):
56 """
57 @param R1_input_file : [str] Path to the R1 fastq file.
58 @param R2_input_file : [str] Path to the R2 fastq file.
59 @param barcode_file : [str] Path to barcodes and samples (one line by sample) description file. Line format : SAMPLE_NAME<TAB>BARCODE.
60 @param mismatches : [int] Number of mismatches allowed
61 @param end : [str] barcode ends ? forward : bol or reverse : eol (def bol)
62 @param global_tmp_files : [str] Path for R1 and R2 files.
63 @param R1_output_files : [list] Paths to the R1 fastq files (one by sample). User provides an empty list.
64 @param R2_output_files : [list] Paths to the R2 fastq files (one by sample). User provides an empty list.
65 @param demultiplex_err_files : [list] Paths to the files with ambiguous and unmatched reads. User provides an empty list.
66 """
67
68 tmp_files = TmpFiles( global_tmp_files.tmp_dir )
69
70 tmp_folder = os.path.join( global_tmp_files.tmp_dir, global_tmp_files.prefix + "_tmp", tmp_files.prefix )
71 global_tmp_files.dirs.append(tmp_folder)
72 if not os.path.exists(tmp_folder):
73 os.mkdir(tmp_folder)
74 self.samples_names = list()
75 # Update output data
76 FH_barcode = open( barcode_file )
77 for line in FH_barcode:
78 sample_name, barcode = line.strip().rsplit(None, 1)
79 # OI DEBUG
80 print "Line is " + line
81 print "Sample name is " + sample_name
82 print "Barcode is " + barcode
83 print "Append to R1_output_files file: " + tmp_folder + " " + sample_name + " " + '_R1.fastq'
84 R1_output_files.append( os.path.join(tmp_folder, sample_name + '_R1.fastq') )
85
86 global_tmp_files.files.append(os.path.join(tmp_folder, sample_name + '_R1.fastq') )
87 if R2_input_file != None:
88 R2_output_files.append( os.path.join(tmp_folder, sample_name + '_R2.fastq') )
89 global_tmp_files.files.append(os.path.join(tmp_folder, sample_name + '_R2.fastq'))
90 self.samples_names.append( sample_name.replace(' ', '_') )
91 FH_barcode.close()
92 self.R1_input_file = R1_input_file
93 self.ambiguous = os.path.join(tmp_folder, 'ambiguous_R1.fastq')
94 self.unmatched = os.path.join(tmp_folder, 'unmatched_R1.fastq')
95 demultiplex_err_files1.extend( [self.ambiguous,self.unmatched] )
96 global_tmp_files.files.extend( [self.ambiguous,self.unmatched] )
97 if R2_input_file != None:
98 demultiplex_err_files2.extend( [os.path.join(tmp_folder, 'ambiguous_R2.fastq'),os.path.join(tmp_folder, 'unmatched_R2.fastq') ])
99 global_tmp_files.files.extend( [os.path.join(tmp_folder, 'ambiguous_R2.fastq'),os.path.join(tmp_folder, 'unmatched_R2.fastq') ])
100
101 # Set class
102 if R2_input_file != None:
103 Cmd.__init__( self,
104 'perl ' + os.environ['TOOL_DIRECTORY'] + '/splitbc.pl',
105 'Demultiplex reads.',
106 R1_input_file + ' ' + R2_input_file + ' --' + end + ' --bcfile ' + barcode_file + ' --mismatches ' + `mismatches` + ' --trim --no_adapt --prefix-r1 ' + os.path.join(tmp_folder, '%_R1.fastq') +\
107 ' --prefix-r2 ' + os.path.join(tmp_folder, '%_R2.fastq') + ' >> ' + demultiplex_log,
108 None )
109 else:
110 Cmd.__init__( self,
111 'perl '+ os.environ['TOOL_DIRECTORY'] + '/splitbc.pl',
112 'Demultiplex reads.',
113 R1_input_file + ' --' + end + ' --bcfile ' + barcode_file + ' --mismatches ' + `mismatches` + ' --trim --no_adapt --prefix-r1 ' + os.path.join(tmp_folder, '%_R1.fastq') +\
114 ' >> ' + demultiplex_log,
115 None )
116
117 def parser(self, log_file):
118 """
119 @summary : Parse the command results to add information in log_file.
120 @log_file : [str] Path to the sample process log file.
121 """
122 # Parse output
123 nb_seq_before = get_fastq_nb_seq(self.R1_input_file)
124 nb_seq_unmatched = get_fastq_nb_seq(self.unmatched)
125 nb_seq_ambiguous = get_fastq_nb_seq(self.ambiguous)
126 # Write result
127 FH_log = Logger( log_file )
128 FH_log.write( 'Results :\n' )
129 FH_log.write( '\tnb seq before demultiplexing : ' + str(nb_seq_before) + '\n' )
130 FH_log.write( '\tnb seq after process matched : ' + str(nb_seq_before - nb_seq_unmatched) + '\n' )
131 FH_log.write( '\tnb seq after process non-ambiguous : ' + str(nb_seq_before - nb_seq_unmatched - nb_seq_ambiguous) + '\n' )
132 FH_log.close()
133
134 def get_version(self):
135 """
136 @summary : Returns the program version number.
137 @return : version number if this is possible, otherwise this method return 'unknown'.
138 """
139 return Cmd.get_version(self, 'stdout')
140
141
142 class Archive(Cmd):
143 """
144 @summary : Creates an archive with files.
145 """
146 def __init__(self, archived_files, archive_path):
147 """
148 @param archived_files: [list] Files added in final archive.
149 @param archive_path: [str] Path to the new archive.
150 """
151
152 tmp_files=TmpFiles( os.path.dirname(archive_path) )
153 tmp_folder = os.path.join( tmp_files.tmp_dir, tmp_files.prefix)
154 tmp_files.dirs.append(tmp_folder)
155 if not os.path.exists(tmp_folder):
156 os.makedirs(tmp_folder)
157
158 if len(archived_files) == 0:
159 raise Exception( "At least one file must be add to the archive '" + archive_path + "'." )
160
161 archived_basenames = list()
162 for current in archived_files:
163 if not os.path.dirname(current) == tmp_folder:
164 os.rename(current, os.path.join(tmp_folder,os.path.basename(current)))
165 tmp_files.files.append(os.path.join(tmp_folder,os.path.basename(current)))
166 archived_basenames.append(os.path.basename(current))
167 # OI DEBUG
168 archive_path_present = "archive_path is file or directory: " + str(os.path.isfile(archive_path)) + " or " + str(os.path.isdir(archive_path))
169 print archive_path_present
170 tmp_folder_present = "tmp_folder is file or directory: " + str(os.path.isfile(tmp_folder)) + " or " + str(os.path.isdir(tmp_folder))
171 print tmp_folder_present
172 for files in archived_basenames:
173 print "archived basename is " + files
174 files_present = "archived basename is file or directory: " + str(os.path.isfile(files)) + " or " + str(os.path.isdir(files))
175 print files_present
176 # END OI DEBUG
177
178
179
180 Cmd.__init__( self,
181 'tar',
182 'Archives files.',
183 '-zcf ' + archive_path + ' -C ' + tmp_folder + " " + " ".join(archived_basenames),
184 None )
185
186 self.Files=tmp_files
187
188
189 def parser(self,log_file):
190 self.Files.deleteAll()
191
192
193 ##################################################################################################################################################
194 #
195 # FUNCTIONS
196 #
197 ##################################################################################################################################################
198 def is_gzip( file ):
199 """
200 @return: [bool] True if the file is gziped.
201 @param file : [str] Path to processed file.
202 """
203 is_gzip = None
204 FH_input = gzip.open( file )
205 try:
206 FH_input.readline()
207 is_gzip = True
208 except:
209 is_gzip = False
210 finally:
211 FH_input.close()
212 return is_gzip
213
214 def split_barcode_file( barcode_file, barcodes_file_list, global_tmp_files ):
215 """
216 @summary: In case of double multiplexe, split barcode file in one forward and multiple reverse barcode files
217 @param barcode_file: [str] Path to the input barcode file
218 @param barcodes_file_list: [list] List of path to the ouput barcode files
219 @param out_dir: [str] path to the output directory to write barcode files
220 """
221 out_dir = global_tmp_files.tmp_dir
222 barcode_input = open(barcode_file,"r")
223 barcode_dict={}
224 for l in barcode_input.readlines():
225 [s,f,r]=l.strip().split()
226 if not "forward_bc" in barcode_dict:
227 barcode_dict["forward_bc"] = [f+"\t"+f]
228 elif not f+"\t"+f in barcode_dict["forward_bc"]:
229 barcode_dict["forward_bc"].append( f+"\t"+f)
230 if not f+"_reverse_bc" in barcode_dict:
231 barcode_dict[f+"_reverse_bc"] = [s+"\t"+r]
232 else :
233 barcode_dict[f+"_reverse_bc"].append(s+"\t"+r)
234
235 f=barcode_dict.pop("forward_bc")
236 barcodes_file_list.append(os.path.join(out_dir,"forward_bc"))
237 global_tmp_files.files.append(os.path.join(out_dir,"forward_bc"))
238 FH_out = open(os.path.join(out_dir,"forward_bc"),"w")
239 FH_out.write("\n".join(f)+"\n")
240 FH_out.close()
241
242 for bc_file in barcode_dict:
243 barcodes_file_list.append(os.path.join(out_dir,bc_file))
244 global_tmp_files.files.append(os.path.join(out_dir,bc_file))
245 FH_out = open(os.path.join(out_dir,bc_file),"w")
246 FH_out.write("\n".join(barcode_dict[bc_file])+"\n")
247 FH_out.close()
248
249 def get_fastq_nb_seq( fastq_file ):
250 """
251 @summary: Returns the number of sequences in fastq_file.
252 @param fastq_file: [str] Path to the fastq file processed.
253 @return: [int] The number of sequences.
254 """
255 FH_input = None
256 if not is_gzip(fastq_file):
257 FH_input = open( fastq_file )
258 else:
259 FH_input = gzip.open( fastq_file )
260 nb_line = 0
261 for line in FH_input:
262 nb_line += 1
263 FH_input.close()
264 nb_seq = nb_line/4
265 return nb_seq
266
267 def concat_files(list_input, output_file):
268
269 FH_out=open(output_file,"w")
270 for f in list_input :
271 FH_in = open(f)
272 string=""
273 i=0
274 for line in FH_in:
275 string+= line
276 i+=1
277 if i==2000 :
278 FH_out.write(string)
279 string=""
280 i=0
281 if i != 0:
282 FH_out.write(string)
283 FH_in.close()
284 FH_out.close()
285
286 def summarise_results( summary_file, barcode_file, log_file ):
287 """
288 @summary: Writes one summary of results from several logs.
289 @param summary_file: [str] The output file.
290 @param log_files: [list] The list of path to log files (one log file by sample).
291 """
292 sample_dict=dict()
293 FH_barcode= open(barcode_file)
294 for line in FH_barcode:
295 sample_dict[line.split()[0]]=0
296
297 FH_summary = open(summary_file, "w")
298 FH_summary.write( "#sample\tcount\n")
299 FH_log = open(log_file,"r")
300 sample_dict["unmatched"]=0
301 sample_dict["ambiguous"]=0
302
303 for line in FH_log.readlines():
304 if line.startswith("Barcode") or line.startswith("total") :
305 pass
306 else :
307 l=line.replace('(','\t').split()
308 if l[0] in sample_dict:
309 sample_dict[l[0]] += int(l[1])
310
311 for s in sample_dict:
312 FH_summary.write(s + '\t' + str(sample_dict[s]) + '\n')
313 FH_summary.close()
314
315
316 ##################################################################################################################################################
317 #
318 # MAIN
319 #
320 ##################################################################################################################################################
321 if __name__ == "__main__":
322 # Manage parameters
323 parser = argparse.ArgumentParser(
324 description='Split by samples the reads in function of inner barcode.'
325 )
326 parser.add_argument('-m', '--mismatches', type=int, default=0, help="Number of mismatches allowed in barcode. [Default: %(default)s]")
327 parser.add_argument('-e', '--end', type=str, default="bol", help="barcode is at the begining of the forward end (bol) or of the reverse (eol) or both (both). [Default: %(default)s]")
328 parser.add_argument( '--debug', default=False, action='store_true', help="Keep temporary files to debug program." )
329 parser.add_argument( '-v', '--version', action='version', version=__version__ )
330 # Inputs
331 group_input = parser.add_argument_group( 'Inputs' )
332 group_input.add_argument( '--input-R1', required=True, help='The R1 sequence file with all samples (format: fastq).' )
333 group_input.add_argument( '--input-R2', default=None, help='The R2 sequence file with all samples (format: fastq).' )
334 group_input.add_argument( '--input-barcode', help='This file describes barcodes and samples (one line by sample). Line format : SAMPLE_NAME<TAB>BARCODE or SAMPLE_NAME<TAB>BARCODE_FW<TAB>BARCODE_RV.' )
335 group_output = parser.add_argument_group( 'Outputs' )
336 # Outputs
337 group_output.add_argument( '--output-demultiplexed', default="demultiplexed_read.tar.gz", help='The tar file containing R1 files and R2 files for each sample (format: tar). [Default: %(default)s]' )
338 group_output.add_argument( '--output-excluded', default="undemultiplexed_read.tar.gz", help='The tar file containing R1 files and R2 files not demultiplexed (format: tar). [Default: %(default)s]' )
339 group_output.add_argument( '-s', '--summary', default='summary.tsv', help='TSV file with summary of filters results (format: TSV). [Default: %(default)s]')
340 group_output.add_argument( '-l', '--log-file', default=sys.stdout, help='This output file will contain several information on executed commands.')
341 args = parser.parse_args()
342 prevent_shell_injections(args)
343
344 Logger.static_write(args.log_file, "## Application\nSoftware :" + sys.argv[0] + " (version : " + str(__version__) + ")\nCommand : " + " ".join(sys.argv) + "\n\n")
345
346 # Process
347 R1_files = list()
348 R2_files = list()
349 tmp_barcode_files = list()
350 tmp_R1_files = list()
351 tmp_R2_files = list()
352 demultiplex_err_files1 = list()
353 demultiplex_err_files2 = list()
354 excluded_R1_file = os.path.join(os.path.split(args.output_demultiplexed)[0],os.path.basename(args.input_R1)+"_excluded_demult")
355 if args.input_R2 != None :
356 excluded_R2_file = os.path.join(os.path.split(args.output_demultiplexed)[0],os.path.basename(args.input_R2)+"_excluded_demult")
357 uniq_id = str(time.time()) + "_" + str(os.getpid())
358
359 tmp_files = TmpFiles( os.path.split(args.output_demultiplexed)[0] )
360 demultiplex_log = tmp_files.add("Demult.log")
361 tmp_folder=tmp_files.add_dir("tmp")
362 os.mkdir(tmp_folder)
363
364 sample_list=[]
365 try:
366 # Process
367 if args.end == "bol" or args.end == "eol" :
368
369 info="\n#Demultiplexing " + os.path.basename(args.input_R1)
370 if args.input_R2 != None:
371 info+= " and " + os.path.basename(args.input_R2)
372 info += " with " + os.path.basename(args.input_barcode) + " in " + args.end + " strand\n"
373 Logger.static_write(args.log_file,info)
374 Demultiplex(args.input_R1, args.input_R2, args.input_barcode, args.mismatches, args.end, tmp_files, R1_files, R2_files, demultiplex_err_files1,demultiplex_err_files2, demultiplex_log).submit( args.log_file )
375 # OI DEBUG
376 for r1file in R1_files:
377 Logger.static_write(args.log_file,"\n\t#Archive demultiplexed files is " + r1file + " \n")
378
379 else:
380 split_barcode_file(args.input_barcode, tmp_barcode_files, tmp_files)
381 info="\n#Demultiplexing " + os.path.basename(args.input_R1)
382 if args.input_R2 != None:
383 info+= " and " + os.path.basename(args.input_R2)
384 info += " with " + os.path.basename(tmp_barcode_files[0]) + " in bol strand\n"
385 Logger.static_write(args.log_file,info)
386 Demultiplex(args.input_R1, args.input_R2, tmp_barcode_files[0], args.mismatches, "bol", tmp_files, tmp_R1_files, tmp_R2_files, demultiplex_err_files1,demultiplex_err_files2, demultiplex_log).submit( args.log_file )
387 for idx,read1_file in enumerate(tmp_R1_files):
388 bc = os.path.basename(read1_file).replace("_R1.fastq","")
389 if os.path.join(tmp_files.tmp_dir,bc+"_reverse_bc") in tmp_barcode_files:
390 if os.stat(tmp_R1_files[idx]).st_size != 0 :
391 info="\n#Demultiplexing " + os.path.basename(tmp_R1_files[idx])
392 if args.input_R2 != None:
393 info+= " and " + os.path.basename(tmp_R2_files[idx])
394 info += " with " + bc+"_reverse_bc" + " in eol strand\n"
395 Logger.static_write(args.log_file,info)
396 if args.input_R2 != None:
397 Demultiplex(tmp_R1_files[idx], tmp_R2_files[idx], os.path.join(tmp_files.tmp_dir,bc+"_reverse_bc"), args.mismatches, "eol", tmp_files, R1_files, R2_files, demultiplex_err_files1, demultiplex_err_files2, demultiplex_log).submit( args.log_file )
398 else:
399 Demultiplex(tmp_R1_files[idx], None, os.path.join(tmp_files.tmp_dir,bc+"_reverse_bc"), args.mismatches, "eol", tmp_files, R1_files, R2_files, demultiplex_err_files1, demultiplex_err_files2, demultiplex_log).submit( args.log_file )
400
401 Logger.static_write(args.log_file,"\n#Summarising result\n")
402 summarise_results( args.summary, args.input_barcode, demultiplex_log )
403 Logger.static_write(args.log_file,"\n#Concatenation of undemultiplexed files 1\n")
404 concat_files(demultiplex_err_files1, excluded_R1_file )
405 if len(R2_files) > 0:
406 Logger.static_write(args.log_file,"\n#Concatenation of undemultiplexed files 2\n")
407 concat_files(demultiplex_err_files2, excluded_R2_file )
408 Logger.static_write(args.log_file,"\n#Archive demultiplexed R1 and R2 files\n")
409 Archive(R1_files + R2_files, args.output_demultiplexed).submit( args.log_file )
410 Logger.static_write(args.log_file,"\n#Archive undemultiplexed R1 and R2 files\n")
411 Archive([excluded_R1_file,excluded_R2_file], args.output_excluded).submit( args.log_file )
412 else:
413 Logger.static_write(args.log_file,"\n#Archive demultiplexed files\n")
414 Archive(R1_files, args.output_demultiplexed).submit( args.log_file )
415 Logger.static_write(args.log_file,"\n#Archive undemultiplexed files\n")
416 Logger.static_write(args.log_file,"\n#Excluded_R1_file is " + excluded_R1_file + "\n")
417 Archive([excluded_R1_file], args.output_excluded).submit( args.log_file )
418
419 # Remove temporary files
420 finally:
421 #OI DEBUG
422 if not args.debug:
423 pass
424 #Logger.static_write(args.log_file,"\n#Removing temporary files\n")
425 #tmp_files.deleteAll()