0
|
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 R1_output_files.append( os.path.join(tmp_folder, sample_name + '_R1.fastq') )
|
|
80 global_tmp_files.files.append(os.path.join(tmp_folder, sample_name + '_R1.fastq') )
|
|
81 if R2_input_file != None:
|
|
82 R2_output_files.append( os.path.join(tmp_folder, sample_name + '_R2.fastq') )
|
|
83 global_tmp_files.files.append(os.path.join(tmp_folder, sample_name + '_R2.fastq'))
|
|
84 self.samples_names.append( sample_name.replace(' ', '_') )
|
|
85 FH_barcode.close()
|
|
86 self.R1_input_file = R1_input_file
|
|
87 self.ambiguous = os.path.join(tmp_folder, 'ambiguous_R1.fastq')
|
|
88 self.unmatched = os.path.join(tmp_folder, 'unmatched_R1.fastq')
|
|
89 demultiplex_err_files1.extend( [self.ambiguous,self.unmatched] )
|
|
90 global_tmp_files.files.extend( [self.ambiguous,self.unmatched] )
|
|
91 if R2_input_file != None:
|
|
92 demultiplex_err_files2.extend( [os.path.join(tmp_folder, 'ambiguous_R2.fastq'),os.path.join(tmp_folder, 'unmatched_R2.fastq') ])
|
|
93 global_tmp_files.files.extend( [os.path.join(tmp_folder, 'ambiguous_R2.fastq'),os.path.join(tmp_folder, 'unmatched_R2.fastq') ])
|
|
94
|
|
95 # Set class
|
|
96 if R2_input_file != None:
|
|
97 Cmd.__init__( self,
|
|
98 'perl splitbc.pl',
|
|
99 'Demultiplex reads.',
|
|
100 R1_input_file + ' ' + R2_input_file + ' --' + end + ' --bcfile ' + barcode_file + ' --mismatches ' + `mismatches` + ' --trim --no_adapt --prefix-r1 ' + os.path.join(tmp_folder, '%_R1.fastq') +\
|
|
101 ' --prefix-r2 ' + os.path.join(tmp_folder, '%_R2.fastq') + ' >> ' + demultiplex_log,
|
|
102 None )
|
|
103 else:
|
|
104 Cmd.__init__( self,
|
|
105 'perl splitbc.pl',
|
|
106 'Demultiplex reads.',
|
|
107 R1_input_file + ' --' + end + ' --bcfile ' + barcode_file + ' --mismatches ' + `mismatches` + ' --trim --no_adapt --prefix-r1 ' + os.path.join(tmp_folder, '%_R1.fastq') +\
|
|
108 ' >> ' + demultiplex_log,
|
|
109 None )
|
|
110
|
|
111 def parser(self, log_file):
|
|
112 """
|
|
113 @summary : Parse the command results to add information in log_file.
|
|
114 @log_file : [str] Path to the sample process log file.
|
|
115 """
|
|
116 # Parse output
|
|
117 nb_seq_before = get_fastq_nb_seq(self.R1_input_file)
|
|
118 nb_seq_unmatched = get_fastq_nb_seq(self.unmatched)
|
|
119 nb_seq_ambiguous = get_fastq_nb_seq(self.ambiguous)
|
|
120 # Write result
|
|
121 FH_log = Logger( log_file )
|
|
122 FH_log.write( 'Results :\n' )
|
|
123 FH_log.write( '\tnb seq before demultiplexing : ' + str(nb_seq_before) + '\n' )
|
|
124 FH_log.write( '\tnb seq after process matched : ' + str(nb_seq_before - nb_seq_unmatched) + '\n' )
|
|
125 FH_log.write( '\tnb seq after process non-ambiguous : ' + str(nb_seq_before - nb_seq_unmatched - nb_seq_ambiguous) + '\n' )
|
|
126 FH_log.close()
|
|
127
|
|
128 def get_version(self):
|
|
129 """
|
|
130 @summary : Returns the program version number.
|
|
131 @return : version number if this is possible, otherwise this method return 'unknown'.
|
|
132 """
|
|
133 return Cmd.get_version(self, 'stdout')
|
|
134
|
|
135
|
|
136 class Archive(Cmd):
|
|
137 """
|
|
138 @summary : Creates an archive with files.
|
|
139 """
|
|
140 def __init__(self, archived_files, archive_path):
|
|
141 """
|
|
142 @param archived_files: [list] Files added in final archive.
|
|
143 @param archive_path: [str] Path to the new archive.
|
|
144 """
|
|
145
|
|
146 tmp_files=TmpFiles( os.path.dirname(archive_path) )
|
|
147 tmp_folder = os.path.join( tmp_files.tmp_dir, tmp_files.prefix)
|
|
148 tmp_files.dirs.append(tmp_folder)
|
|
149 if not os.path.exists(tmp_folder):
|
|
150 os.makedirs(tmp_folder)
|
|
151
|
|
152 if len(archived_files) == 0:
|
|
153 raise Exception( "At least one file must be add to the archive '" + archive_path + "'." )
|
|
154
|
|
155 archived_basenames = list()
|
|
156 for current in archived_files:
|
|
157 if not os.path.dirname(current) == tmp_folder:
|
|
158 os.rename(current, os.path.join(tmp_folder,os.path.basename(current)))
|
|
159 tmp_files.files.append(os.path.join(tmp_folder,os.path.basename(current)))
|
|
160 archived_basenames.append(os.path.basename(current))
|
|
161
|
|
162
|
|
163 Cmd.__init__( self,
|
|
164 'tar',
|
|
165 'Archives files.',
|
|
166 '-zcf ' + archive_path + ' -C ' + tmp_folder + " " + " ".join(archived_basenames),
|
|
167 None )
|
|
168
|
|
169 self.Files=tmp_files
|
|
170
|
|
171
|
|
172 def parser(self,log_file):
|
|
173 self.Files.deleteAll()
|
|
174
|
|
175
|
|
176 ##################################################################################################################################################
|
|
177 #
|
|
178 # FUNCTIONS
|
|
179 #
|
|
180 ##################################################################################################################################################
|
|
181 def is_gzip( file ):
|
|
182 """
|
|
183 @return: [bool] True if the file is gziped.
|
|
184 @param file : [str] Path to processed file.
|
|
185 """
|
|
186 is_gzip = None
|
|
187 FH_input = gzip.open( file )
|
|
188 try:
|
|
189 FH_input.readline()
|
|
190 is_gzip = True
|
|
191 except:
|
|
192 is_gzip = False
|
|
193 finally:
|
|
194 FH_input.close()
|
|
195 return is_gzip
|
|
196
|
|
197 def split_barcode_file( barcode_file, barcodes_file_list, global_tmp_files ):
|
|
198 """
|
|
199 @summary: In case of double multiplexe, split barcode file in one forward and multiple reverse barcode files
|
|
200 @param barcode_file: [str] Path to the input barcode file
|
|
201 @param barcodes_file_list: [list] List of path to the ouput barcode files
|
|
202 @param out_dir: [str] path to the output directory to write barcode files
|
|
203 """
|
|
204 out_dir = global_tmp_files.tmp_dir
|
|
205 barcode_input = open(barcode_file,"r")
|
|
206 barcode_dict={}
|
|
207 for l in barcode_input.readlines():
|
|
208 [s,f,r]=l.strip().split()
|
|
209 if not "forward_bc" in barcode_dict:
|
|
210 barcode_dict["forward_bc"] = [f+"\t"+f]
|
|
211 elif not f+"\t"+f in barcode_dict["forward_bc"]:
|
|
212 barcode_dict["forward_bc"].append( f+"\t"+f)
|
|
213 if not f+"_reverse_bc" in barcode_dict:
|
|
214 barcode_dict[f+"_reverse_bc"] = [s+"\t"+r]
|
|
215 else :
|
|
216 barcode_dict[f+"_reverse_bc"].append(s+"\t"+r)
|
|
217
|
|
218 f=barcode_dict.pop("forward_bc")
|
|
219 barcodes_file_list.append(os.path.join(out_dir,"forward_bc"))
|
|
220 global_tmp_files.files.append(os.path.join(out_dir,"forward_bc"))
|
|
221 FH_out = open(os.path.join(out_dir,"forward_bc"),"w")
|
|
222 FH_out.write("\n".join(f)+"\n")
|
|
223 FH_out.close()
|
|
224
|
|
225 for bc_file in barcode_dict:
|
|
226 barcodes_file_list.append(os.path.join(out_dir,bc_file))
|
|
227 global_tmp_files.files.append(os.path.join(out_dir,bc_file))
|
|
228 FH_out = open(os.path.join(out_dir,bc_file),"w")
|
|
229 FH_out.write("\n".join(barcode_dict[bc_file])+"\n")
|
|
230 FH_out.close()
|
|
231
|
|
232 def get_fastq_nb_seq( fastq_file ):
|
|
233 """
|
|
234 @summary: Returns the number of sequences in fastq_file.
|
|
235 @param fastq_file: [str] Path to the fastq file processed.
|
|
236 @return: [int] The number of sequences.
|
|
237 """
|
|
238 FH_input = None
|
|
239 if not is_gzip(fastq_file):
|
|
240 FH_input = open( fastq_file )
|
|
241 else:
|
|
242 FH_input = gzip.open( fastq_file )
|
|
243 nb_line = 0
|
|
244 for line in FH_input:
|
|
245 nb_line += 1
|
|
246 FH_input.close()
|
|
247 nb_seq = nb_line/4
|
|
248 return nb_seq
|
|
249
|
|
250 def concat_files(list_input, output_file):
|
|
251
|
|
252 FH_out=open(output_file,"w")
|
|
253 for f in list_input :
|
|
254 FH_in = open(f)
|
|
255 string=""
|
|
256 i=0
|
|
257 for line in FH_in:
|
|
258 string+= line
|
|
259 i+=1
|
|
260 if i==2000 :
|
|
261 FH_out.write(string)
|
|
262 string=""
|
|
263 i=0
|
|
264 if i != 0:
|
|
265 FH_out.write(string)
|
|
266 FH_in.close()
|
|
267 FH_out.close()
|
|
268
|
|
269 def summarise_results( summary_file, barcode_file, log_file ):
|
|
270 """
|
|
271 @summary: Writes one summary of results from several logs.
|
|
272 @param summary_file: [str] The output file.
|
|
273 @param log_files: [list] The list of path to log files (one log file by sample).
|
|
274 """
|
|
275 sample_dict=dict()
|
|
276 FH_barcode= open(barcode_file)
|
|
277 for line in FH_barcode:
|
|
278 sample_dict[line.split()[0]]=0
|
|
279
|
|
280 FH_summary = open(summary_file, "w")
|
|
281 FH_summary.write( "#sample\tcount\n")
|
|
282 FH_log = open(log_file,"r")
|
|
283 sample_dict["unmatched"]=0
|
|
284 sample_dict["ambiguous"]=0
|
|
285
|
|
286 for line in FH_log.readlines():
|
|
287 if line.startswith("Barcode") or line.startswith("total") :
|
|
288 pass
|
|
289 else :
|
|
290 l=line.replace('(','\t').split()
|
|
291 if l[0] in sample_dict:
|
|
292 sample_dict[l[0]] += int(l[1])
|
|
293
|
|
294 for s in sample_dict:
|
|
295 FH_summary.write(s + '\t' + str(sample_dict[s]) + '\n')
|
|
296 FH_summary.close()
|
|
297
|
|
298
|
|
299 ##################################################################################################################################################
|
|
300 #
|
|
301 # MAIN
|
|
302 #
|
|
303 ##################################################################################################################################################
|
|
304 if __name__ == "__main__":
|
|
305 # Manage parameters
|
|
306 parser = argparse.ArgumentParser(
|
|
307 description='Split by samples the reads in function of inner barcode.'
|
|
308 )
|
|
309 parser.add_argument('-m', '--mismatches', type=int, default=0, help="Number of mismatches allowed in barcode. [Default: %(default)s]")
|
|
310 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]")
|
|
311 parser.add_argument( '--debug', default=False, action='store_true', help="Keep temporary files to debug program." )
|
|
312 parser.add_argument( '-v', '--version', action='version', version=__version__ )
|
|
313 # Inputs
|
|
314 group_input = parser.add_argument_group( 'Inputs' )
|
|
315 group_input.add_argument( '--input-R1', required=True, help='The R1 sequence file with all samples (format: fastq).' )
|
|
316 group_input.add_argument( '--input-R2', default=None, help='The R2 sequence file with all samples (format: fastq).' )
|
|
317 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.' )
|
|
318 group_output = parser.add_argument_group( 'Outputs' )
|
|
319 # Outputs
|
|
320 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]' )
|
|
321 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]' )
|
|
322 group_output.add_argument( '-s', '--summary', default='summary.tsv', help='TSV file with summary of filters results (format: TSV). [Default: %(default)s]')
|
|
323 group_output.add_argument( '-l', '--log-file', default=sys.stdout, help='This output file will contain several information on executed commands.')
|
|
324 args = parser.parse_args()
|
|
325 prevent_shell_injections(args)
|
|
326
|
|
327 Logger.static_write(args.log_file, "## Application\nSoftware :" + sys.argv[0] + " (version : " + str(__version__) + ")\nCommand : " + " ".join(sys.argv) + "\n\n")
|
|
328
|
|
329 # Process
|
|
330 R1_files = list()
|
|
331 R2_files = list()
|
|
332 tmp_barcode_files = list()
|
|
333 tmp_R1_files = list()
|
|
334 tmp_R2_files = list()
|
|
335 demultiplex_err_files1 = list()
|
|
336 demultiplex_err_files2 = list()
|
|
337 excluded_R1_file = os.path.join(os.path.split(args.output_demultiplexed)[0],os.path.basename(args.input_R1)+"_excluded_demult")
|
|
338 if args.input_R2 != None :
|
|
339 excluded_R2_file = os.path.join(os.path.split(args.output_demultiplexed)[0],os.path.basename(args.input_R2)+"_excluded_demult")
|
|
340 uniq_id = str(time.time()) + "_" + str(os.getpid())
|
|
341
|
|
342 tmp_files = TmpFiles( os.path.split(args.output_demultiplexed)[0] )
|
|
343 demultiplex_log = tmp_files.add("Demult.log")
|
|
344 tmp_folder=tmp_files.add_dir("tmp")
|
|
345 os.mkdir(tmp_folder)
|
|
346
|
|
347 sample_list=[]
|
|
348 try:
|
|
349 # Process
|
|
350 if args.end == "bol" or args.end == "eol" :
|
|
351
|
|
352 info="\n#Demultiplexing " + os.path.basename(args.input_R1)
|
|
353 if args.input_R2 != None:
|
|
354 info+= " and " + os.path.basename(args.input_R2)
|
|
355 info += " with " + os.path.basename(args.input_barcode) + " in " + args.end + " strand\n"
|
|
356 Logger.static_write(args.log_file,info)
|
|
357 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 )
|
|
358 else:
|
|
359 split_barcode_file(args.input_barcode, tmp_barcode_files, tmp_files)
|
|
360 info="\n#Demultiplexing " + os.path.basename(args.input_R1)
|
|
361 if args.input_R2 != None:
|
|
362 info+= " and " + os.path.basename(args.input_R2)
|
|
363 info += " with " + os.path.basename(tmp_barcode_files[0]) + " in bol strand\n"
|
|
364 Logger.static_write(args.log_file,info)
|
|
365 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 )
|
|
366 for idx,read1_file in enumerate(tmp_R1_files):
|
|
367 bc = os.path.basename(read1_file).replace("_R1.fastq","")
|
|
368 if os.path.join(tmp_files.tmp_dir,bc+"_reverse_bc") in tmp_barcode_files:
|
|
369 if os.stat(tmp_R1_files[idx]).st_size != 0 :
|
|
370 info="\n#Demultiplexing " + os.path.basename(tmp_R1_files[idx])
|
|
371 if args.input_R2 != None:
|
|
372 info+= " and " + os.path.basename(tmp_R2_files[idx])
|
|
373 info += " with " + bc+"_reverse_bc" + " in eol strand\n"
|
|
374 Logger.static_write(args.log_file,info)
|
|
375 if args.input_R2 != None:
|
|
376 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 )
|
|
377 else:
|
|
378 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 )
|
|
379
|
|
380 Logger.static_write(args.log_file,"\n#Summarising result\n")
|
|
381 summarise_results( args.summary, args.input_barcode, demultiplex_log )
|
|
382
|
|
383 Logger.static_write(args.log_file,"\n#Concatenation of undemultiplexed files 1\n")
|
|
384 concat_files(demultiplex_err_files1, excluded_R1_file )
|
|
385 if len(R2_files) > 0:
|
|
386 Logger.static_write(args.log_file,"\n#Concatenation of undemultiplexed files 2\n")
|
|
387 concat_files(demultiplex_err_files2, excluded_R2_file )
|
|
388 Logger.static_write(args.log_file,"\n#Archive demultiplexed R1 and R2 files\n")
|
|
389 Archive(R1_files + R2_files, args.output_demultiplexed).submit( args.log_file )
|
|
390 Logger.static_write(args.log_file,"\n#Archive undemultiplexed R1 and R2 files\n")
|
|
391 Archive([excluded_R1_file,excluded_R2_file], args.output_excluded).submit( args.log_file )
|
|
392 else:
|
|
393 Logger.static_write(args.log_file,"\n#Archive demultiplexed files\n")
|
|
394 Archive(R1_files, args.output_demultiplexed).submit( args.log_file )
|
|
395 Logger.static_write(args.log_file,"\n#Archive undemultiplexed files\n")
|
|
396 Archive([excluded_R1_file], args.output_excluded).submit( args.log_file )
|
|
397
|
|
398 # Remove temporary files
|
|
399 finally:
|
|
400 if not args.debug:
|
|
401 Logger.static_write(args.log_file,"\n#Removing temporary files\n")
|
|
402 tmp_files.deleteAll()
|
|
403
|