13 my $CONFIDENCE = 0.95;
16 my $NMB = 1024; # default
20 my $read_type = 1; # default, single end with qual
55 my $keep_intermediate_files = 0;
57 my $strand_specific = 0;
60 my ($time_start, $time_end, $time_alignment, $time_rsem, $time_ci) = (0, 0, 0, 0, 0);
62 GetOptions("keep-intermediate-files" => \$keep_intermediate_files,
63 "no-qualities" => \$no_qual,
64 "paired-end" => \$paired_end,
65 "strand-specific" => \$strand_specific,
68 "sam-header-info=s" => \$fn_list,
70 "seed-length=i" => \$L,
71 "bowtie-path=s" => \$bowtie_path,
74 "bowtie-m=i" => \$maxHits,
75 "phred33-quals" => \$phred33,
76 "phred64-quals" => \$phred64, #solexa1.3-quals" => \$phred64,
77 "solexa-quals" => \$solexa,
78 "forward-prob=f" => \$probF,
79 "fragment-length-min=i" => \$minL,
80 "fragment-length-max=i" => \$maxL,
81 "fragment-length-mean=f" => \$mean,
82 "fragment-length-sd=f" => \$sd,
83 "estimate-rspd" => \$estRSPD,
84 "num-rspd-bins=i" => \$B,
85 "p|num-threads=i" => \$nThreads,
86 "out-bam" => \$genBamF,
87 "sampling-for-bam" => \$sampling,
88 "calc-ci" => \$calcCI,
89 "ci-memory=i" => \$NMB,
92 "h|help" => \$help) or pod2usage(-exitval => 2, -verbose => 2);
94 pod2usage(-verbose => 2) if ($help == 1);
97 #check parameters and options
99 if ($is_sam || $is_bam) {
100 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (scalar(@ARGV) != 3);
101 pod2usage(-msg => "--sam and --bam cannot be active at the same time!", -exitval => 2, -verbose => 2) if ($is_sam == 1&& $is_bam == 1);
102 pod2usage(-msg => "--bowtie-path, --bowtie-n, --bowtie-e, --bowtie-m, --phred33-quals, --phred64-quals or --solexa-quals cannot be set if input is SAM/BAM format!", -exitval => 2, -verbose => 2) if ($bowtie_path ne "" || $C != 2 || $E != 99999999 || $maxHits != 200 || $phred33 || $phred64 || $solexa);
105 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (!$paired_end && scalar(@ARGV) != 3 || $paired_end && scalar(@ARGV) != 4);
106 pod2usage(-msg => "Only one of --phred33-quals --phred64-quals/--solexa1.3-quals --solexa-suqls can be active!", -exitval => 2, -verbose => 2) if ($phred33 + $phred64 + $solexa > 1);
107 podwusage(-msg => "--sam , --bam or --sam-header-info cannot be set if use bowtie aligner to produce alignments!", -exitval => 2, -verbose => 2) if ($is_sam || $is_bam || $fn_list ne "");
110 pod2usage(-msg => "Forward probability should be in [0, 1]!", -exitval => 2, -verbose => 2) if ($probF < 0 || $probF > 1);
111 pod2usage(-msg => "Min fragment length should be at least 1!", -exitval => 2, -verbose => 2) if ($minL < 1);
112 pod2usage(-msg => "Min fragment length should be smaller or equal to max fragment length!", -exitval => 2, -verbose => 2) if ($minL > $maxL);
113 pod2usage(-msg => "The memory allocated for calculating credibility intervals should be at least 1 MB!\n", -exitval => 2, -verbose => 2) if ($NMB < 1);
114 pod2usage(-msg => "Number of threads should be at least 1!\n", -exitval => 2, -verbose => 2) if ($nThreads < 1);
115 pod2usage(-msg => "Seed length should be at least 25!\n", -exitval => 2, -verbose => 2) if ($L < 25);
116 pod2usage(-msg => "--sampling-for-bam cannot be specified if --out-bam is not specified!\n", -exitval => 2, -verbose => 2) if ($sampling && !$genBamF);
118 if ($strand_specific) { $probF = 1.0; }
124 my ($refName, $sampleName, $sampleToken, $temp_dir, $stat_dir, $imdName) = ();
128 if ($no_qual) { $read_type = 2; }
129 else { $read_type = 3; }
132 if ($no_qual) { $read_type = 0; }
133 else { $read_type = 1; }
136 if (scalar(@ARGV) == 3) {
137 if ($is_sam || $is_bam) { $inpF = $ARGV[0]; }
138 else {$mate1_list = $ARGV[0]; }
140 $sampleName = $ARGV[2];
143 $mate1_list = $ARGV[0];
144 $mate2_list = $ARGV[1];
146 $sampleName = $ARGV[3];
149 my $pos = rindex($sampleName, '/');
150 if ($pos < 0) { $sampleToken = $sampleName; }
151 else { $sampleToken = substr($sampleName, $pos + 1); }
153 $temp_dir = "$sampleName.temp";
154 $stat_dir = "$sampleName.stat";
156 if (!(-d $temp_dir) && !mkdir($temp_dir)) { print "Fail to create folder $temp_dir.\n"; exit(-1); }
157 if (!(-d $stat_dir) && !mkdir($stat_dir)) { print "Fail to create folder $stat_dir.\n"; exit(-1); }
159 $imdName = "$temp_dir/$sampleToken";
161 if (!$is_sam && !$is_bam && $phred33 + $phred64 + $solexa == 0) { $phred33 = 1; }
163 my ($mate_minL, $mate_maxL) = (1, $maxL);
165 if ($bowtie_path ne "") { $bowtie_path .= "/"; }
167 my ($fn, $dir, $suf) = fileparse($0);
170 if (!$is_sam && !$is_bam) {
171 $command = $bowtie_path."bowtie";
172 if ($read_type == 0 || $read_type == 2) { $command .= " -f"; }
173 else { $command .= " -q"; }
175 if ($phred33) { $command .= " --phred33-quals"; }
176 elsif ($phred64) { $command .= " --phred64-quals"; }
177 elsif ($solexa) { $command .= " --solexa-quals"; }
178 else { print "Oh, no!!!"; exit(2); }
180 $command .= " -n $C -e $E -l $L";
182 if ($read_type == 2 || $read_type == 3) { $command .= " -I $minL -X $maxL"; }
184 if ($strand_specific || $probF == 1.0) { $command .= " --norc"; }
185 elsif ($probF == 0.0) { $command .= " --nofw"; }
187 $command .= " -p $nThreads -a -m $maxHits -S";
188 if ($quiet) { $command .= " --quiet"; }
190 $command .= " $refName";
191 if ($read_type == 0 || $read_type == 1) {
192 $command .= " $mate1_list";
195 $command .= " -1 $mate1_list -2 $mate2_list";
198 $command .= " | gzip > $imdName.sam.gz";
201 if ($mTime) { $time_start = time(); }
203 $status = system($command);
205 if ($mTime) { $time_end = time(); $time_alignment = $time_end - $time_start; }
208 print "bowtie failed! Please check if you provide correct parameters/options for the pipeline!\n";
213 $inpF = "$imdName.sam.gz";
214 $is_sam = 1; # output of bowtie is a sam file
217 if ($mTime) { $time_start = time(); }
219 $command = $dir."rsem-parse-alignments $refName $sampleName $sampleToken";
222 if ($is_sam) { $samInpType = "s"; }
223 elsif ($is_bam) { $samInpType = "b"; }
225 $command .= " $samInpType $inpF -t $read_type";
226 if ($fn_list ne "") { $command .= " -l $fn_list"; }
227 if ($tagName ne "") { $command .= " -tag $tagName"; }
228 if ($quiet) { $command .= " -q"; }
231 $status = system($command);
233 print "rsem-parse-alignments failed! Please check if you provide correct parameters/options for the pipeline!\n";
238 $command = $dir."rsem-build-read-index $gap";
240 case 0 { $command .= " 0 $quiet $imdName\_alignable.fa"; }
241 case 1 { $command .= " 1 $quiet $imdName\_alignable.fq"; }
242 case 2 { $command .= " 0 $quiet $imdName\_alignable_1.fa $imdName\_alignable_2.fa"; }
243 case 3 { $command .= " 1 $quiet $imdName\_alignable_1.fq $imdName\_alignable_2.fq"; }
246 $status = system($command);
248 print "rsem-build-read-index failed! Please check if you provide correct parameters/options for the pipeline!\n";
253 my $doesOpen = open(OUTPUT, ">$imdName.mparams");
254 if ($doesOpen == 0) { print "Cannot generate $imdName.mparams!\n"; exit(-1); }
255 print OUTPUT "$minL $maxL\n";
256 print OUTPUT "$probF\n";
257 print OUTPUT "$estRSPD\n";
259 print OUTPUT "$mate_minL $mate_maxL\n";
260 print OUTPUT "$mean $sd\n";
264 $command = $dir."rsem-run-em $refName $read_type $sampleName $sampleToken -p $nThreads";
266 $command .= " -b $samInpType $inpF";
267 if ($fn_list ne "") { $command .= " 1 $fn_list"; }
268 else { $command .= " 0"; }
269 if ($sampling) { $command .= " --sampling"; }
271 if ($calcCI) { $command .= " --gibbs-out"; }
272 if ($quiet) { $command .= " -q"; }
275 $status = system($command);
277 print "rsem-run-em failed! Please check if you provide correct parameters/options for the pipeline!\n";
283 $command = $dir."sam/samtools sort $sampleName.bam $sampleName.sorted";
285 $status = system($command);
287 print "sam/samtools sort failed! Please check if you provide correct parameters/options for the pipeline!\n";
291 $command = $dir."sam/samtools index $sampleName.sorted.bam";
293 $status = system($command);
295 print "sam/samtools index failed! Please check if you provide correct parameters/options for the pipeline!\n";
301 &collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
302 &collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
304 if ($mTime) { $time_end = time(); $time_rsem = $time_end - $time_start; }
306 if ($mTime) { $time_start = time(); }
309 $command = $dir."rsem-run-gibbs $refName $sampleName $sampleToken $BURNIN $CHAINLEN $SAMPLEGAP";
310 # $command .= " -p $nThreads";
311 if ($quiet) { $command .= " -q"; }
313 $status = system($command);
315 print "rsem-run-gibbs failed! Please check if you provide correct parameters/options for the pipeline!\n";
320 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak1");
321 system("mv $sampleName.genes.results $imdName.genes.results.bak1");
322 &collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
323 &collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
325 $command = $dir."rsem-calculate-credibility-intervals $refName $sampleName $sampleToken $CONFIDENCE $NSPC $NMB";
326 if ($quiet) { $command .= " -q"; }
328 $status = system($command);
330 print "rsem-calculate-credibility-intervals failed! Please check if you provide correct parameters/options for the pipeline!\n";
335 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak2");
336 system("mv $sampleName.genes.results $imdName.genes.results.bak2");
337 &collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
338 &collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
341 if ($mTime) { $time_end = time(); $time_ci = $time_end - $time_start; }
343 if ($mTime) { $time_start = time(); }
345 if (!$keep_intermediate_files) {
346 $status = system("rm -rf $temp_dir");
348 print "Fail to delete the temporary folder!\n";
353 if ($mTime) { $time_end = time(); }
356 open(OUTPUT, ">$sampleName.time");
357 print OUTPUT "Alignment: $time_alignment s.\n";
358 print OUTPUT "RSEM: $time_rsem s.\n";
359 print OUTPUT "CI: $time_ci s.\n";
360 my $time_del = $time_end - $time_start;
361 print OUTPUT "Delete: $time_del s.\n";
369 my (@results, @comment) = ();
376 $local_status = open(INPUT, $inpF);
377 if ($local_status == 0) { print "Fail to open file $inpF!\n"; exit(-1); }
382 while ($line = <INPUT>) {
385 my @local_arr = split(/\t/, $line);
386 if ($cnt == 4) { @comment = @local_arr; }
387 else { push(@results, \@local_arr); }
390 push(@results, \@comment);
393 $local_status = open(OUTPUT, ">$outF");
394 if ($local_status == 0) { print "Fail to create file $outF!\n"; exit(-1); }
396 my $n = scalar(@results);
397 my $m = scalar(@{$results[0]});
398 for (my $i = 0; $i < $m; $i++) {
400 for (my $j = 0; $j < $n; $j++) { push(@out_arr, $results[$j][$i]); }
402 print OUTPUT "@out_arr\n";
412 rsem-calculate-expression
418 rsem-calculate-expression [options] upstream_read_file(s) reference_name sample_name
419 rsem-calculate-expression [options] --paired-end upstream_read_file(s) downstream_read_file(s) reference_name sample_name
420 rsem-calculate-expression [options] --sam/--bam [--paired-end] input reference_name sample_name
428 =item B<upstream_read_files(s)>
430 Comma-separated list of files containing single-end reads or upstream reads for paired-end data. By default, these files are assumed to be in FASTQ format. If the --no-qualities option is specified, then FASTA format is expected.
432 =item B<downstream_read_file(s)>
434 Comma-separated list of files containing downstream reads which are paired with the upstream reads. By default, these files are assumed to be in FASTQ format. If the --no-qualities option is specified, then FASTA format is expected.
438 SAM/BAM formatted input file. If "-" is specified for the filename, SAM/BAM input is instead assumed to come from standard input. RSEM requires all alignments of the same read group together. For paired-end reads, RSEM also requires the two mates of any alignment be adjacent. See Description section for how to make input file obey RSEM's requirements.
440 =item B<reference_name>
442 The name of the reference used. The user must have run 'rsem-prepare-reference' with this reference_name before running this program.
446 The name of the sample analyzed. All output files are prefixed by this name (e.g., sample_name.genes.results)
454 =item B<--paired-end>
456 Input reads are paired-end reads. (Default: off)
458 =item B<--no-qualities>
460 Input reads do not contain quality scores. (Default: off)
462 =item B<--strand-specific>
464 The RNA-Seq protocol used to generate the reads is strand specific, i.e., all (upstream) reads are derived from the forward strand. This option is equivalent to --forward-prob=1.0. With this option set, if RSEM runs the Bowtie aligner, the '--norc' Bowtie option will be used, which disables alignment to the reverse strand of transcripts. (Default: off)
468 Input file is in SAM format. (Default: off)
472 Input file is in BAM format. (Default: off)
474 =item B<--sam-header-info> <file>
476 RSEM reads header information from input by default. If this option is on, header information is read from the specified file. For the format of the file, please see SAM official website. (Default: "")
478 =item B<-p/--num-threads> <int>
480 Number of threads to use. Both Bowtie and expression estimation will use this many threads. (Default: 1)
484 Generate a BAM file, 'sample_name.bam', with alignments mapped to genomic coordinates and annotated with their posterior probabilities. In addition, RSEM will call samtools (included in RSEM package) to sort and index the bam file. 'sample_name.sorted.bam' and 'sample_name.sorted.bam.bai' will be generated. (Default: off)
486 =item B<--sampling-for-bam>
488 When RSEM generates a BAM file, instead of outputing all alignments a read has with their posterior probabilities, one alignment is sampled and outputed according to the posterior probabilities. If the sampling result is that the read comes from the "noise" transcript, nothing is outputed. It cannot be specified unless --out-bam is specified. (Default: off)
492 Calculate 95% credibility intervals and posterior mean estimates. (Default: off)
494 =item B<--seed-length> <int>
496 Seed length used by the read aligner. Providing the correct value for this parameter is important for RSEM's accuracy if the data are single-end reads. If RSEM runs Bowtie, it uses this value for Bowtie's seed length parameter. The minimum value is 25. Any read with its or at least one of its mates' (for paired-end reads) length less than 25 will be ignored. (Default: 25)
498 =item B<--tag> <string>
500 The name of the optional field used in the SAM input for identifying a read with too many valid alignments. The field should have the format <tagName>:i:<value>, where a <value> bigger than 0 indicates a read with too many alignments. (Default: "")
502 =item B<--bowtie-path> <path>
504 The path to the bowtie executables. (Default: the path to the bowtie executables is assumed to be in the user's PATH environment variable)
506 =item B<--bowtie-n> <int>
508 (Bowtie parameter) max # of mismatches in the seed. (Range: 0-3, Default: 2)
510 =item B<--bowtie-e> <int>
512 (Bowtie parameter) max sum of mismatch quality scores across the alignment. (Default: 99999999)
514 =item B<--bowtie-m> <int>
516 (Bowtie parameter) suppress all alignments for a read if > <int> valid alignments exist. (Default: 200)
518 =item B<--phred33-quals>
520 Input quality scores are encoded as Phred+33. (Default: on)
522 =item B<--phred64-quals>
524 Input quality scores are encoded as Phred+64 (default for GA Pipeline ver. >= 1.3). (Default: off)
526 =item B<--solexa-quals>
528 Input quality scores are solexa encoded (from GA Pipeline ver. < 1.3). (Default: off)
530 =item B<--forward-prob> <double>
532 Probability of generating a read from the forward strand of a transcript. Set to 1 for a strand-specific protocol where all (upstream) reads are derived from the forward strand, 0 for a strand-specific protocol where all (upstream) read are derived from the reverse strand, or 0.5 for a non-strand-specific protocol. (Default: 0.5)
534 =item B<--fragment-length-min> <int>
536 Minimum read/insert length allowed. This is also the value for the bowtie -I option. (Default: 1)
538 =item B<--fragment-length-max> <int>
540 Maximum read/insert length allowed. This is also the value for the bowtie -X option. (Default: 1000)
542 =item B<--fragment-length-mean> <double>
544 (single-end data only) The mean of the fragment length distribution, which is assumed to be a Gaussian. (Default: -1, which disables use of the fragment length distribution)
546 =item B<--fragment-length-sd> <double>
548 (single-end data only) The standard deviation of the fragment length distribution, which is assumed to be a Gaussian. (Default: 0, which assumes that all fragments are of the same length, given by the rounded value of B<--fragment-length-mean>)
550 =item B<--estimate-rspd>
552 Set this option if you want to estimate the read start position distribution (RSPD) from data. Otherwise, RSEM will use a uniform RSPD. (Default: off)
554 =item B<--num-rspd-bins> <int>
556 Number of bins in the RSPD. Only relevant when '--estimate-rspd' is specified. Use of the default setting is recommended. (Default: 20)
558 =item B<--ci-memory> <int>
560 Amount of memory (in MB) RSEM is allowed to use for computing credibility intervals. (Default: 1024)
562 =item B<--keep-intermediate-files>
564 Keep temporary files generated by RSEM. RSEM creates a temporary directory, 'sample_name.temp', into which it puts all intermediate output files. If this directory already exists, RSEM overwrites all files generated by previous RSEM runs inside of it. By default, after RSEM finishes, the temporary directory is deleted. Set this option to prevent the deletion of this directory and the intermediate files inside of it. (Default: off)
568 Suppress the output of logging information. (Default: off)
572 Show help information.
578 In its default mode, this program aligns input reads against a reference transcriptome with Bowtie and calculates expression values using the alignments. RSEM assumes the data are single-end reads with quality scores, unless the '--paired-end' or '--no-qualities' options are specified. Users may use an alternative aligner by specifying one of the --sam and --bam options, and providing an alignment file in the specified format. However, users should make sure the alignment file satisfies the requirements mentioned in ARGUMENTS section.
580 One simple way to make the alignment file (e.g. input.sam) satisfying RSEM's requirements (assuming the aligner used put mates in a paired-end read adjacent) is to use the following command:
582 sort -k 1,1 -s input.sam > input.sorted.sam
584 The SAM/BAM format RSEM uses is v1.3. However, it is compatible with old SAM/BAM format.
586 The user must run 'rsem-prepare-reference' with the appropriate reference before using this program.
588 For single-end data, it is strongly recommended that the user provide the fragment length distribution parameters (--fragment-length-mean and --fragment-length-sd). For paired-end data, RSEM will automatically learn a fragment length distribution from the data.
590 Please note that some of the default values for the Bowtie parameters are not the same as those defined for Bowtie itself.
592 The temporary directory and all intermediate files will be removed when RSEM finishes unless '--keep-intermediate-files' is specified.
594 With the "--calc-ci" option, 95% credibility intervals and posterior mean estimates will be calculated in addition to maximum likelihood estimates.
600 =item B<sample_name.genes.results>
602 File containing gene level expression estimates. The format of each
603 line in this file is:
605 gene_id expected_counts tau_value [pmc_value tau_pme_value tau_ci_lower_bound tau_ci_upper_bound] transcript_id_list
607 Fields are separated by the tab character. Fields within "[]" are only
608 presented if '--calc-ci' is set. pme stands for posterior mean
609 estimation. pmc stands for posterior mean counts. ci_lower_bound(l)
610 means the lower bound of the credibility intervals, ci_upper_bound(u)
611 means the upper bound of the credibility intervals. So the credibility
612 interval is [l, u]. 'transcript_id_list' is a space-separated list of
613 transcript_ids belonging to the gene.
615 =item B<sample_name.isoforms.results>
617 File containing isoform level expression values. The format of each
618 line in this file is:
620 transcript_id expected_counts tau_value [pmc_value tau_pme_value tau_ci_lower_bound tau_ci_upper_bound] other_attributes
622 Fields are separated by the tab character. 'other_attributes' are all
623 other attributes after attribute 'transcript_id' field in the GTF
624 file. If no other attributes are given or no GTF file is provided in
625 'rsem-prepare-reference', there will be no tab after the
628 =item B<sample_name.bam, sample_name.sorted.bam and sample_name.sorted.bam.bai>
630 Only generated when --out-bam is specified.
632 'sample_name.bam' is a BAM-formatted file of read alignments in
633 genomic coordinates. Alignments of reads that have identical genomic
634 coordinates (i.e., alignments to different isoforms that share the
635 same genomic region) are collapsed into one alignment. The MAPQ field
636 of each alignment is set to min(100, floor(-10 * log10(1.0 - w) +
637 0.5)), where w is the posterior probability of that alignment being
638 the true mapping of a read. In addition, RSEM pads a new tag
639 ZW:f:value, where value is a single precision floating number
640 representing the posterior probability.
642 'sample_name.sorted.bam' and 'sample_name.sorted.bam.bai' are the
643 sorted BAM file and indices generated by samtools (included in RSEM package).
645 =item B<sample_name.stat>
647 This is a folder instead of a file. All model related statistics are stored in this folder. Use 'rsem-plot-model' can generate plots using this folder.
653 Assume the path to the bowtie executables is in the user's PATH environment variable. Reference files are under '/ref' with name 'mm9'.
655 1) '/data/mmliver.fq', single-end reads with quality scores. Quality scores are encoded as for 'GA pipeline version >= 1.3'. We want to use 8 threads and generate a BAM file:
657 rsem-calculate-expression --phred64-quals \
664 2) '/data/mmliver_1.fq' and '/data/mmliver_2.fq', paired-end reads with quality scores. Quality scores are in SANGER format. We want to use 8 threads and do not generate a BAM file:
666 rsem-calculate-expression -p 8 \
671 mmliver_paired_end_quals
673 3) '/data/mmliver.fa', single-end reads without quality scores. We want to use 8 threads and generate a BAM file:
675 rsem-calculate-expression -p 8 \
679 mmliver_single_without_quals
681 4) Data are the same as 1). We want to take a fragment length distribution into consideration. We set the fragment length mean to 150 and the standard deviation to 35. In addition to a BAM file, we also want to generate credibility intervals. We allow RSEM to use 1GB of memory for CI calculation.
683 rsem-calculate-expression --bowtie-path /sw/bowtie \
685 --fragment-length-mean 150.0 \
686 --fragment-length-sd 35.0 \
695 5) '/data/mmliver_paired_end_quals.bam', paired-end reads with quality scores. We want to use 8 threads and do not generate a BAM file:
697 rsem-calculate-expression --paired-end \
700 /data/mmliver_paired_end_quals.bam \
702 mmliver_paired_end_quals