|
0
|
1 #!/usr/bin/env python
|
|
|
2
|
|
|
3 # Supports Cufflinks versions 1.3 and newer (updated based on 2.2.1).
|
|
|
4
|
|
|
5 import optparse, os, shutil, subprocess, sys, tempfile
|
|
|
6 from galaxy import eggs
|
|
|
7 from galaxy.datatypes.util.gff_util import parse_gff_attributes, gff_attributes_to_str
|
|
|
8
|
|
|
9 def stop_err( msg ):
|
|
|
10 sys.stderr.write( "%s\n" % msg )
|
|
|
11 sys.exit()
|
|
|
12
|
|
|
13 def __main__():
|
|
|
14 #Parse Command Line
|
|
|
15 parser = optparse.OptionParser()
|
|
|
16 parser.add_option( '-1', '--input', dest='input', help=' file of RNA-Seq read alignments in the SAM format. SAM is a standard short read alignment, that allows aligners to attach custom tags to individual alignments, and Cufflinks requires that the alignments you supply have some of these tags. Please see Input formats for more details.' )
|
|
|
17 parser.add_option( '-I', '--max-intron-length', dest='max_intron_len', help='The minimum intron length. Cufflinks will not report transcripts with introns longer than this, and will ignore SAM alignments with REF_SKIP CIGAR operations longer than this. The default is 300,000.' )
|
|
|
18 parser.add_option( '-F', '--min-isoform-fraction', dest='min_isoform_fraction', help='After calculating isoform abundance for a gene, Cufflinks filters out transcripts that it believes are very low abundance, because isoforms expressed at extremely low levels often cannot reliably be assembled, and may even be artifacts of incompletely spliced precursors of processed transcripts. This parameter is also used to filter out introns that have far fewer spliced alignments supporting them. The default is 0.05, or 5% of the most abundant isoform (the major isoform) of the gene.' )
|
|
|
19 parser.add_option( '-j', '--pre-mrna-fraction', dest='pre_mrna_fraction', help='Some RNA-Seq protocols produce a significant amount of reads that originate from incompletely spliced transcripts, and these reads can confound the assembly of fully spliced mRNAs. Cufflinks uses this parameter to filter out alignments that lie within the intronic intervals implied by the spliced alignments. The minimum depth of coverage in the intronic region covered by the alignment is divided by the number of spliced reads, and if the result is lower than this parameter value, the intronic alignments are ignored. The default is 5%.' )
|
|
|
20 parser.add_option( '-p', '--num-threads', dest='num_threads', help='Use this many threads to align reads. The default is 1.' )
|
|
|
21 parser.add_option( '-G', '--GTF', dest='GTF', help='Tells Cufflinks to use the supplied reference annotation to estimate isoform expression. It will not assemble novel transcripts, and the program will ignore alignments not structurally compatible with any reference transcript.' )
|
|
|
22 parser.add_option ("--compatible-hits-norm",dest='compatible_hits_norm',help='Count hits compatible with reference RNAs only')
|
|
|
23 parser.add_option( '-g', '--GTF-guide', dest='GTFguide', help='use reference transcript annotation to guide assembly' )
|
|
|
24 parser.add_option("--3-overhang-tolerance",dest='3_overhang_tolerance', help='The number of bp allowed to overhang the 3prime end of a reference transcript when determining if an assembled transcript should be merged with it (ie, the assembled transcript is not novel). The default is 600 bp.')
|
|
|
25 parser.add_option("--intron-overhang-tolerance",dest='intron_overhang_tolerance',help='The number of bp allowed to enter the intron of a reference transcript when determining if an assembled transcript should be merged with it (ie, the assembled transcript is not novel). The default is 50 bp.')
|
|
|
26 parser.add_option("--no-faux-reads", dest='no_faux_reads',help='This option disables tiling of the reference transcripts with faux reads. Use this if you only want to use sequencing reads in assembly but do not want to output assembled transcripts that lay within reference transcripts. All reference transcripts in the input annotation will also be included in the output.')
|
|
|
27 parser.add_option( '-u', '--multi-read-correct', dest='multi_read_correct', action="store_true", help='Tells Cufflinks to do an initial estimation procedure to more accurately weight reads mapping to multiple locations in the genome')
|
|
|
28
|
|
|
29 # Normalization options.
|
|
|
30 parser.add_option( "--no-effective-length-correction", dest="no_effective_length_correction", action="store_true" )
|
|
|
31 parser.add_option( "--no-length-correction", dest="no_length_correction", action="store_true" )
|
|
|
32
|
|
|
33 # Wrapper / Galaxy options.
|
|
|
34 parser.add_option( '-A', '--assembled-isoforms-output', dest='assembled_isoforms_output_file', help='Assembled isoforms output file; formate is GTF.' )
|
|
|
35
|
|
|
36 # Advanced Options:
|
|
|
37 parser.add_option( "--library-type",dest="library_type",help=' library prep used for input reads, default fr-unstranded')
|
|
|
38 parser.add_option( '-M','--mask-file', dest='mask_file', help='Tells Cufflinks to ignore all reads that could have come from transcripts in this GTF file. \
|
|
|
39 We recommend including any annotated rRNA, mitochondrial transcripts other abundant transcripts \
|
|
|
40 you wish to ignore in your analysis in this file. Due to variable efficiency of mRNA enrichment \
|
|
|
41 methods and rRNA depletion kits, masking these transcripts often improves the overall robustness \
|
|
|
42 of transcript abundance estimates.')
|
|
|
43 parser.add_option( '-m', '--inner-mean-dist', dest='inner_mean_dist', help='This is the expected (mean) inner distance between mate pairs. \
|
|
|
44 For, example, for paired end runs with fragments selected at 300bp, \
|
|
|
45 where each end is 50bp, you should set -r to be 200. The default is 45bp.') # cufflinks: --frag-len-mean
|
|
|
46
|
|
|
47 parser.add_option( '-s', '--inner-dist-std-dev', dest='inner_dist_std_dev', help='The standard deviation for the distribution on inner distances between mate pairs. The default is 20bp.' ) # cufflinks: --frag-len-std-dev
|
|
|
48 parser.add_option( '--max-mle-iterations', dest='max_mle_iterations', help='Sets the number of iterations allowed during maximum likelihood estimation of abundances. Default: 5000' )
|
|
|
49 parser.add_option( '--junc-alpha', dest='junc_alpha', help='Alpha value for the binomial test used during false positive spliced alignment filtration. Default: 0.001' )
|
|
|
50 parser.add_option( '--small-anchor-fraction', dest='small_anchor_fraction', help='Spliced reads with less than this percent of their length on each side of\
|
|
|
51 the junction are considered suspicious and are candidates for filtering prior to assembly. Default: 0.09.' )
|
|
|
52 parser.add_option( '--overhang-tolerance', dest='overhang_tolerance', help='The number of bp allowed to enter the intron of a transcript when determining if a \
|
|
|
53 read or another transcript is mappable to/compatible with it. The default is 8 bp based on the default bowtie/TopHat parameters.' )
|
|
|
54 parser.add_option( '--max-bundle-length', dest='max_bundle_length', help='Maximum genomic length of a given bundle" help="Default: 3,500,000bp' )
|
|
|
55 parser.add_option( '--max-bundle-frags', dest='max_bundle_frags', help='Sets the maximum number of fragments a locus may have before being skipped. Skipped loci are listed in skipped.gtf. Default: 1,000,000' )
|
|
|
56 parser.add_option( '--min-intron-length', dest='min_intron_length', help='Minimal allowed intron size. Default: 50' )
|
|
|
57 parser.add_option( '--trim-3-avgcov-thresh', dest='trim_3_avgcov_thresh', help='Minimum average coverage required to attempt 3prime trimming. Default: 10' )
|
|
|
58 parser.add_option( '--trim-3-dropoff-frac', dest='trim_3_dropoff_frac', help='The fraction of average coverage below which to trim the 3prime end of an assembled transcript. Default: 0.1' )
|
|
|
59
|
|
|
60
|
|
|
61 # Bias correction options.
|
|
|
62 parser.add_option( '-b', dest='do_bias_correction', action="store_true", help='Providing Cufflinks with a multifasta file via this option instructs it to run our new bias detection and correction algorithm which can significantly improve accuracy of transcript abundance estimates.')
|
|
|
63 parser.add_option( '', '--index', dest='index', help='The path of the reference genome' )
|
|
|
64 parser.add_option( '', '--ref_file', dest='ref_file', help='The reference dataset from the history' )
|
|
|
65
|
|
|
66 # Global model (for trackster).
|
|
|
67 parser.add_option( '', '--global_model', dest='global_model_file', help='Global model used for computing on local data' )
|
|
|
68
|
|
|
69 (options, args) = parser.parse_args()
|
|
|
70
|
|
|
71 # output version # of tool
|
|
|
72 try:
|
|
|
73 tmp = tempfile.NamedTemporaryFile().name
|
|
|
74 tmp_stdout = open( tmp, 'wb' )
|
|
|
75 proc = subprocess.Popen( args='cufflinks --no-update-check 2>&1', shell=True, stdout=tmp_stdout )
|
|
|
76 tmp_stdout.close()
|
|
|
77 returncode = proc.wait()
|
|
|
78 stdout = None
|
|
|
79 for line in open( tmp_stdout.name, 'rb' ):
|
|
|
80 if line.lower().find( 'cufflinks v' ) >= 0:
|
|
|
81 stdout = line.strip()
|
|
|
82 break
|
|
|
83 if stdout:
|
|
|
84 sys.stdout.write( '%s\n' % stdout )
|
|
|
85 else:
|
|
|
86 raise Exception
|
|
|
87 except:
|
|
|
88 sys.stdout.write( 'Could not determine Cufflinks version\n' )
|
|
|
89
|
|
|
90 # If doing bias correction, set/link to sequence file.
|
|
|
91 if options.do_bias_correction:
|
|
|
92 if options.ref_file:
|
|
|
93 # Sequence data from history.
|
|
|
94 # Create symbolic link to ref_file so that index will be created in working directory.
|
|
|
95 seq_path = "ref.fa"
|
|
|
96 os.symlink( options.ref_file, seq_path )
|
|
|
97 else:
|
|
|
98 if not os.path.exists( options.index ):
|
|
|
99 stop_err( 'Reference genome %s not present, request it by reporting this error.' % options.index )
|
|
|
100 seq_path = options.index
|
|
|
101
|
|
|
102 # Build command.
|
|
|
103
|
|
|
104 # Base; always use quiet mode to avoid problems with storing log output.
|
|
|
105 cmd = "cufflinks -q --no-update-check"
|
|
|
106
|
|
|
107 # Add options.
|
|
|
108 if options.max_intron_len:
|
|
|
109 cmd += ( " -I %i" % int ( options.max_intron_len ) )
|
|
|
110 if options.min_isoform_fraction:
|
|
|
111 cmd += ( " -F %f" % float ( options.min_isoform_fraction ) )
|
|
|
112 if options.pre_mrna_fraction:
|
|
|
113 cmd += ( " -j %f" % float ( options.pre_mrna_fraction ) )
|
|
|
114 if options.num_threads:
|
|
|
115 cmd += ( " -p %i" % int ( options.num_threads ) )
|
|
|
116 if options.GTF:
|
|
|
117 cmd += ( " -G %s" % options.GTF )
|
|
|
118 if options.compatible_hits_norm:
|
|
|
119 cmd += ( " --compatible-hits-norm" )
|
|
|
120 if options.GTFguide:
|
|
|
121 cmd += ( " -g %s" % options.GTFguide )
|
|
|
122 cmd += ( " --3-overhang-tolerance %i" % int ( options.3_overhang_tolerance ) )
|
|
|
123 cmd += ( " --intron-overhang-tolerance %i" % int ( options.intron_overhang_tolerance ) )
|
|
|
124 if options.no_faux_reads:
|
|
|
125 cmd += ( " --no-faux-reads" )
|
|
|
126 if options.multi_read_correct:
|
|
|
127 cmd += ( " -u" )
|
|
|
128
|
|
|
129 if options.library_type && options.library_type != 'auto':
|
|
|
130 cmd += ( " --library-type %s" % options.library_type)
|
|
|
131 if options.mask_file:
|
|
|
132 cmd += ( " --mask-file %s" % options.mask_file )
|
|
|
133 if options.inner_mean_dist:
|
|
|
134 cmd += ( " -m %i" % int ( options.inner_mean_dist ) )
|
|
|
135 if options.inner_dist_std_dev:
|
|
|
136 cmd += ( " -s %i" % int ( options.inner_dist_std_dev ) )
|
|
|
137 if options.max_mle_iterations:
|
|
|
138 cmd += ( " --max-mle-iterations %i" % int ( options.max_mle_iterations ) )
|
|
|
139 if options.junc_alpha:
|
|
|
140 cmd += ( " --junc-alpha %f" % float ( options.junc_alpha) )
|
|
|
141 if options.small_anchor_fraction:
|
|
|
142 cmd += ( " --small-anchor-fraction %f" % float (options.small_anchor_fraction ) )
|
|
|
143 if options.overhang_tolerance:
|
|
|
144 cmd += ( " --overhang-tolerance %i" % int ( options.overhang_tolerance ) )
|
|
|
145 if options.max_bundle_length:
|
|
|
146 cmd += ( " --max-bundle-length %i" % int ( options.max_bundle_length ) )
|
|
|
147 if options.max_bundle_frags:
|
|
|
148 cmd += ( " --max-bundle-frags %i" % int ( options.max_bundle_frags ) )
|
|
|
149 if options.min_intron_length:
|
|
|
150 cmd += ( " --min-intron-length %i" % int ( options.min_intron_length ) )
|
|
|
151 if options.trim_3_avgcov_thresh:
|
|
|
152 cmd += ( " --trim-3-avgcov-thresh %i" % int ( options.trim_3_avgcov_thresh ) )
|
|
|
153 if options.trim_3_dropoff_frac:
|
|
|
154 cmd += ( " --trim-3-dropoff-frac %f" % float ( options.trim_3_dropoff_frac ) )
|
|
|
155
|
|
|
156 if options.do_bias_correction:
|
|
|
157 cmd += ( " -b %s" % seq_path )
|
|
|
158 if options.no_effective_length_correction:
|
|
|
159 cmd += ( " --no-effective-length-correction" )
|
|
|
160 if options.no_length_correction:
|
|
|
161 cmd += ( " --no-length-correction" )
|
|
|
162
|
|
|
163 # Add input files.
|
|
|
164 cmd += " " + options.input
|
|
|
165
|
|
|
166 # Debugging.
|
|
|
167 print cmd
|
|
|
168
|
|
|
169 #
|
|
|
170 # Run command and handle output.
|
|
|
171 #
|
|
|
172 try:
|
|
|
173 #
|
|
|
174 # Run command.
|
|
|
175 #
|
|
|
176 tmp_name = tempfile.NamedTemporaryFile( dir="." ).name
|
|
|
177 tmp_stderr = open( tmp_name, 'wb' )
|
|
|
178 proc = subprocess.Popen( args=cmd, shell=True, stderr=tmp_stderr.fileno() )
|
|
|
179 returncode = proc.wait()
|
|
|
180 tmp_stderr.close()
|
|
|
181
|
|
|
182 # Error checking.
|
|
|
183 if returncode != 0:
|
|
|
184 raise Exception, "return code = %i" % returncode
|
|
|
185
|
|
|
186 #
|
|
|
187 # Handle output.
|
|
|
188 #
|
|
|
189
|
|
|
190 # Read standard error to get total map/upper quartile mass.
|
|
|
191 total_map_mass = -1
|
|
|
192 tmp_stderr = open( tmp_name, 'r' )
|
|
|
193 for line in tmp_stderr:
|
|
|
194 if line.lower().find( "map mass" ) >= 0 or line.lower().find( "upper quartile" ) >= 0:
|
|
|
195 total_map_mass = float( line.split(":")[1].strip() )
|
|
|
196 break
|
|
|
197 tmp_stderr.close()
|
|
|
198
|
|
|
199 #
|
|
|
200 # If there's a global model provided, use model's total map mass
|
|
|
201 # to adjust FPKM + confidence intervals.
|
|
|
202 #
|
|
|
203 if options.global_model_file:
|
|
|
204 # Global model is simply total map mass from original run.
|
|
|
205 global_model_file = open( options.global_model_file, 'r' )
|
|
|
206 global_model_total_map_mass = float( global_model_file.readline() )
|
|
|
207 global_model_file.close()
|
|
|
208
|
|
|
209 # Ratio of global model's total map mass to original run's map mass is
|
|
|
210 # factor used to adjust FPKM.
|
|
|
211 fpkm_map_mass_ratio = total_map_mass / global_model_total_map_mass
|
|
|
212
|
|
|
213 # Update FPKM values in transcripts.gtf file.
|
|
|
214 transcripts_file = open( "transcripts.gtf", 'r' )
|
|
|
215 tmp_transcripts = tempfile.NamedTemporaryFile( dir="." ).name
|
|
|
216 new_transcripts_file = open( tmp_transcripts, 'w' )
|
|
|
217 for line in transcripts_file:
|
|
|
218 fields = line.split( '\t' )
|
|
|
219 attrs = parse_gff_attributes( fields[8] )
|
|
|
220 attrs[ "FPKM" ] = str( float( attrs[ "FPKM" ] ) * fpkm_map_mass_ratio )
|
|
|
221 attrs[ "conf_lo" ] = str( float( attrs[ "conf_lo" ] ) * fpkm_map_mass_ratio )
|
|
|
222 attrs[ "conf_hi" ] = str( float( attrs[ "conf_hi" ] ) * fpkm_map_mass_ratio )
|
|
|
223 fields[8] = gff_attributes_to_str( attrs, "GTF" )
|
|
|
224 new_transcripts_file.write( "%s\n" % '\t'.join( fields ) )
|
|
|
225 transcripts_file.close()
|
|
|
226 new_transcripts_file.close()
|
|
|
227 shutil.copyfile( tmp_transcripts, "transcripts.gtf" )
|
|
|
228
|
|
|
229 # TODO: update expression files as well.
|
|
|
230
|
|
|
231 # Set outputs. Transcript and gene expression handled by wrapper directives.
|
|
|
232 shutil.copyfile( "transcripts.gtf" , options.assembled_isoforms_output_file )
|
|
|
233 if total_map_mass > -1:
|
|
|
234 f = open( "global_model.txt", 'w' )
|
|
|
235 f.write( "%f\n" % total_map_mass )
|
|
|
236 f.close()
|
|
|
237 except Exception, e:
|
|
|
238 # Read stderr so that it can be reported:
|
|
|
239 tmp_stderr = open( tmp_name, 'rb' )
|
|
|
240 stderr = ''
|
|
|
241 buffsize = 1048576
|
|
|
242 try:
|
|
|
243 while True:
|
|
|
244 stderr += tmp_stderr.read( buffsize )
|
|
|
245 if not stderr or len( stderr ) % buffsize != 0:
|
|
|
246 break
|
|
|
247 except OverflowError:
|
|
|
248 pass
|
|
|
249 tmp_stderr.close()
|
|
|
250
|
|
|
251 stop_err( 'Error running cufflinks.\n%s\n%s' % ( str( e ), stderr ) )
|
|
|
252
|
|
|
253 if __name__=="__main__": __main__()
|
