13 my $CONFIDENCE = 0.95;
16 my $NMB = 1024; # default
20 my $read_type = 1; # default, single end with qual
54 my $keep_intermediate_files = 0;
56 my $strand_specific = 0;
58 GetOptions("keep-intermediate-files" => \$keep_intermediate_files,
59 "no-qualities" => \$no_qual,
60 "paired-end" => \$paired_end,
61 "strand-specific" => \$strand_specific,
64 "sam-header-info=s" => \$fn_list,
66 "seed-length=i" => \$L,
67 "bowtie-path=s" => \$bowtie_path,
70 "bowtie-m=i" => \$maxHits,
71 "phred33-quals" => \$phred33,
72 "phred64-quals" => \$phred64, #solexa1.3-quals" => \$phred64,
73 "solexa-quals" => \$solexa,
74 "forward-prob=f" => \$probF,
75 "fragment-length-min=i" => \$minL,
76 "fragment-length-max=i" => \$maxL,
77 "fragment-length-mean=f" => \$mean,
78 "fragment-length-sd=f" => \$sd,
79 "estimate-rspd" => \$estRSPD,
80 "num-rspd-bins=i" => \$B,
81 "p|num-threads=i" => \$nThreads,
82 "out-bam" => \$genBamF,
83 "calc-ci" => \$calcCI,
84 "ci-memory=i" => \$NMB,
86 "h|help" => \$help) or pod2usage(-exitval => 2, -verbose => 2);
88 pod2usage(-verbose => 2) if ($help == 1);
91 #check parameters and options
93 if ($is_sam || $is_bam) {
94 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (scalar(@ARGV) != 3);
95 pod2usage(-msg => "--sam and --bam cannot be active at the same time!", -exitval => 2, -verbose => 2) if ($is_sam == 1&& $is_bam == 1);
96 pod2usage(-msg => "--bowtie-path, --bowtie-n, --bowtie-e, --bowtie-m, --phred33-quals, --phred64-quals or --solexa-quals cannot be set if input is SAM/BAM format!", -exitval => 2, -verbose => 2) if ($bowtie_path ne "" || $C != 2 || $E != 99999999 || $maxHits != 200 || $phred33 || $phred64 || $solexa);
99 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (!$paired_end && scalar(@ARGV) != 3 || $paired_end && scalar(@ARGV) != 4);
100 pod2usage(-msg => "Only one of --phred33-quals --phred64-quals/--solexa1.3-quals --solexa-suqls can be active!", -exitval => 2, -verbose => 2) if ($phred33 + $phred64 + $solexa > 1);
101 podwusage(-msg => "--sam , --bam or --sam-header-info cannot be set if use bowtie aligner to produce alignments!", -exitval => 2, -verbose => 2) if ($is_sam || $is_bam || $fn_list ne "");
104 pod2usage(-msg => "Forward probability should be in [0, 1]!", -exitval => 2, -verbose => 2) if ($probF < 0 || $probF > 1);
105 pod2usage(-msg => "Min fragment length should be at least 1!", -exitval => 2, -verbose => 2) if ($minL < 1);
106 pod2usage(-msg => "Min fragment length should be smaller or equal to max fragment length!", -exitval => 2, -verbose => 2) if ($minL > $maxL);
107 pod2usage(-msg => "The memory allocated for calculating credibility intervals should be at least 1 MB!\n", -exitval => 2, -verbose => 2) if ($NMB < 1);
108 pod2usage(-msg => "Number of threads should be at least 1!\n", -exitval => 2, -verbose => 2) if ($nThreads < 1);
110 if ($strand_specific) { $probF = 1.0; }
116 my ($refName, $sampleName, $sampleToken, $temp_dir, $stat_dir, $imdName) = ();
120 if ($no_qual) { $read_type = 2; }
121 else { $read_type = 3; }
124 if ($no_qual) { $read_type = 0; }
125 else { $read_type = 1; }
128 if (scalar(@ARGV) == 3) {
129 if ($is_sam || $is_bam) { $inpF = $ARGV[0]; }
130 else {$mate1_list = $ARGV[0]; }
132 $sampleName = $ARGV[2];
135 $mate1_list = $ARGV[0];
136 $mate2_list = $ARGV[1];
138 $sampleName = $ARGV[3];
141 my $pos = rindex($sampleName, '/');
142 if ($pos < 0) { $sampleToken = $sampleName; }
143 else { $sampleToken = substr($sampleName, $pos + 1); }
145 $temp_dir = "$sampleName.temp";
146 $stat_dir = "$sampleName.stat";
148 if (!(-d $temp_dir) && !mkdir($temp_dir)) { print "Fail to create folder $temp_dir.\n"; exit(-1); }
149 if (!(-d $stat_dir) && !mkdir($stat_dir)) { print "Fail to create folder $stat_dir.\n"; exit(-1); }
151 $imdName = "$temp_dir/$sampleToken";
153 if (!$is_sam && !$is_bam && $phred33 + $phred64 + $solexa == 0) { $phred33 = 1; }
155 my ($mate_minL, $mate_maxL) = (1, $maxL);
157 if ($bowtie_path ne "") { $bowtie_path .= "/"; }
159 my ($fn, $dir, $suf) = fileparse($0);
162 if (!$is_sam && !$is_bam) {
163 $command = $bowtie_path."bowtie";
164 if ($read_type == 0 || $read_type == 2) { $command .= " -f"; }
165 else { $command .= " -q"; }
167 if ($phred33) { $command .= " --phred33-quals"; }
168 elsif ($phred64) { $command .= " --phred64-quals"; }
169 elsif ($solexa) { $command .= " --solexa-quals"; }
170 else { print "Oh, no!!!"; exit(2); }
172 $command .= " -n $C -e $E -l $L";
174 if ($read_type == 2 || $read_type == 3) { $command .= " -I $minL -X $maxL"; }
176 if ($strand_specific || $probF == 1.0) { $command .= " --norc"; }
177 elsif ($probF == 0.0) { $command .= " --nofw"; }
179 $command .= " -p $nThreads -a -m $maxHits -S";
180 if ($quiet) { $command .= " --quiet"; }
182 $command .= " $refName";
183 if ($read_type == 0 || $read_type == 1) {
184 $command .= " $mate1_list";
187 $command .= " -1 $mate1_list -2 $mate2_list";
190 $command .= " | gzip > $imdName.sam.gz";
192 $status = system($command);
194 print "bowtie failed! Please check if you provide correct parameters/options for the pipeline!\n";
199 $inpF = "$imdName.sam.gz";
200 $is_sam = 1; # output of bowtie is a sam file
203 $command = $dir."rsem-parse-alignments $refName $sampleName $sampleToken";
206 if ($is_sam) { $samInpType = "s"; }
207 elsif ($is_bam) { $samInpType = "b"; }
209 $command .= " $samInpType $inpF -t $read_type";
210 if ($fn_list ne "") { $command .= " -l $fn_list"; }
211 if ($tagName ne "") { $command .= " -tag $tagName"; }
212 if ($quiet) { $command .= " -q"; }
215 $status = system($command);
217 print "rsem-parse-alignments failed! Please check if you provide correct parameters/options for the pipeline!\n";
222 $command = $dir."rsem-build-read-index $gap";
224 case 0 { $command .= " 0 $quiet $imdName\_alignable.fa"; }
225 case 1 { $command .= " 1 $quiet $imdName\_alignable.fq"; }
226 case 2 { $command .= " 0 $quiet $imdName\_alignable_1.fa $imdName\_alignable_2.fa"; }
227 case 3 { $command .= " 1 $quiet $imdName\_alignable_1.fq $imdName\_alignable_2.fq"; }
230 $status = system($command);
232 print "rsem-build-read-index failed! Please check if you provide correct parameters/options for the pipeline!\n";
237 $status = open(OUTPUT, ">$imdName.mparams");
238 if ($status == 0) { print "Cannot generate $imdName.mparams!\n"; exit(-1); }
239 print OUTPUT "$minL $maxL\n";
240 print OUTPUT "$probF\n";
241 print OUTPUT "$estRSPD\n";
243 print OUTPUT "$mate_minL $mate_maxL\n";
244 print OUTPUT "$mean $sd\n";
248 $command = $dir."rsem-run-em $refName $read_type $sampleName $sampleToken -p $nThreads";
250 $command .= " -b $samInpType $inpF";
251 if ($fn_list ne "") { $command .= " 1 $fn_list"; }
252 else { $command .= " 0"; }
254 if ($calcCI) { $command .= " --gibbs-out"; }
255 if ($quiet) { $command .= " -q"; }
258 $status = system($command);
260 print "rsem-run-em failed! Please check if you provide correct parameters/options for the pipeline!\n";
266 $command = $dir."sam/samtools sort $sampleName.bam $sampleName.sorted";
268 $status = system($command);
270 print "sam/samtools sort failed! Please check if you provide correct parameters/options for the pipeline!\n";
274 $command = $dir."sam/samtools index $sampleName.sorted.bam";
276 $status = system($command);
278 print "sam/samtools index failed! Please check if you provide correct parameters/options for the pipeline!\n";
284 &collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
285 &collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
288 $command = $dir."rsem-run-gibbs $refName $sampleName $sampleToken $BURNIN $CHAINLEN $SAMPLEGAP";
289 # $command .= " -p $nThreads";
290 if ($quiet) { $command .= " -q"; }
292 $status = system($command);
294 print "rsem-run-gibbs failed! Please check if you provide correct parameters/options for the pipeline!\n";
299 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak1");
300 system("mv $sampleName.genes.results $imdName.genes.results.bak1");
301 &collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
302 &collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
304 $command = $dir."rsem-calculate-credibility-intervals $refName $sampleName $sampleToken $CONFIDENCE $NSPC $NMB";
305 if ($quiet) { $command .= " -q"; }
307 $status = system($command);
309 print "rsem-calculate-credibility-intervals failed! Please check if you provide correct parameters/options for the pipeline!\n";
314 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak2");
315 system("mv $sampleName.genes.results $imdName.genes.results.bak2");
316 &collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
317 &collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
320 if (!$keep_intermediate_files) {
321 $status = system ("rm -rf $temp_dir");
323 print "Fail to delete the temporary folder!\n";
332 my (@results, @comment) = ();
339 $local_status = open(INPUT, $inpF);
340 if ($local_status == 0) { print "Fail to open file $inpF!\n"; exit(-1); }
345 while ($line = <INPUT>) {
348 my @local_arr = split(/\t/, $line);
349 if ($cnt == 4) { @comment = @local_arr; }
350 else { push(@results, \@local_arr); }
353 push(@results, \@comment);
356 $local_status = open(OUTPUT, ">$outF");
357 if ($local_status == 0) { print "Fail to create file $outF!\n"; exit(-1); }
359 my $n = scalar(@results);
360 my $m = scalar(@{$results[0]});
361 for (my $i = 0; $i < $m; $i++) {
363 for (my $j = 0; $j < $n; $j++) { push(@out_arr, $results[$j][$i]); }
365 print OUTPUT "@out_arr\n";
375 rsem-calculate-expression
381 rsem-calculate-expression [options] upstream_read_file(s) reference_name sample_name
382 rsem-calculate-expression [options] --paired-end upstream_read_file(s) downstream_read_file(s) reference_name sample_name
383 rsem-calculate-expression [options] --sam/--bam [--paired-end] input reference_name sample_name
391 =item B<upstream_read_files(s)>
393 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.
395 =item B<downstream_read_file(s)>
397 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.
401 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.
403 =item B<reference_name>
405 The name of the reference used. The user must have run 'rsem-prepare-reference' with this reference_name before running this program.
409 The name of the sample analyzed. All output files are prefixed by this name (e.g., sample_name.genes.results)
417 =item B<--paired-end>
419 Input reads are paired-end reads. (Default: off)
421 =item B<--no-qualities>
423 Input reads do not contain quality scores. (Default: off)
425 =item B<--strand-specific>
427 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)
431 Input file is in SAM format. (Default: off)
435 Input file is in BAM format. (Default: off)
437 =item B<--sam-header-info> <file>
439 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: "")
441 =item B<-p/--num-threads> <int>
443 Number of threads to use. Both Bowtie and expression estimation will use this many threads. (Default: 1)
447 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)
451 Calculate 95% credibility intervals and posterior mean estimates. (Default: off)
453 =item B<--seed-length> <int>
455 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. (Default: 25)
457 =item B<--tag> <string>
459 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: "")
461 =item B<--bowtie-path> <path>
463 The path to the bowtie executables. (Default: the path to the bowtie executables is assumed to be in the user's PATH environment variable)
465 =item B<--bowtie-n> <int>
467 (Bowtie parameter) max # of mismatches in the seed. (Range: 0-3, Default: 2)
469 =item B<--bowtie-e> <int>
471 (Bowtie parameter) max sum of mismatch quality scores across the alignment. (Default: 99999999)
473 =item B<--bowtie-m> <int>
475 (Bowtie parameter) suppress all alignments for a read if > <int> valid alignments exist. (Default: 200)
477 =item B<--phred33-quals>
479 Input quality scores are encoded as Phred+33. (Default: on)
481 =item B<--phred64-quals>
483 Input quality scores are encoded as Phred+64 (default for GA Pipeline ver. >= 1.3). (Default: off)
485 =item B<--solexa-quals>
487 Input quality scores are solexa encoded (from GA Pipeline ver. < 1.3). (Default: off)
489 =item B<--forward-prob> <double>
491 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)
493 =item B<--fragment-length-min> <int>
495 Minimum read/insert length allowed. This is also the value for the bowtie -I option. (Default: 1)
497 =item B<--fragment-length-max> <int>
499 Maximum read/insert length allowed. This is also the value for the bowtie -X option. (Default: 1000)
501 =item B<--fragment-length-mean> <double>
503 (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)
505 =item B<--fragment-length-sd> <double>
507 (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>)
509 =item B<--estimate-rspd>
511 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)
513 =item B<--num-rspd-bins> <int>
515 Number of bins in the RSPD. Only relevant when '--estimate-rspd' is specified. Use of the default setting is recommended. (Default: 20)
517 =item B<--ci-memory> <int>
519 Amount of memory (in MB) RSEM is allowed to use for computing credibility intervals. (Default: 1024)
521 =item B<--keep-intermediate-files>
523 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)
527 Suppress the output of logging information. (Default: off)
531 Show help information.
537 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.
539 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:
541 sort -k 1,1 -s input.sam > input.sorted.sam
543 The SAM/BAM format RSEM uses is v1.3. However, it is compatible with old SAM/BAM format.
545 The user must run 'rsem-prepare-reference' with the appropriate reference before using this program.
547 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.
549 Please note that some of the default values for the Bowtie parameters are not the same as those defined for Bowtie itself.
551 The temporary directory and all intermediate files will be removed when RSEM finishes unless '--keep-intermediate-files' is specified.
553 With the "--calc-ci" option, 95% credibility intervals and posterior mean estimates will be calculated in addition to maximum likelihood estimates.
559 =item B<sample_name.genes.results>
561 File containing gene level expression estimates. The format of each
562 line in this file is:
564 gene_id expected_counts tau_value [pmc_value tau_pme_value tau_ci_lower_bound tau_ci_upper_bound] transcript_id_list
566 Fields are separated by the tab character. Fields within "[]" are only
567 presented if '--calc-ci' is set. pme stands for posterior mean
568 estimation. pmc stands for posterior mean counts. ci_lower_bound(l)
569 means the lower bound of the credibility intervals, ci_upper_bound(u)
570 means the upper bound of the credibility intervals. So the credibility
571 interval is [l, u]. 'transcript_id_list' is a space-separated list of
572 transcript_ids belonging to the gene.
574 =item B<sample_name.isoforms.results>
576 File containing isoform level expression values. The format of each
577 line in this file is:
579 transcript_id expected_counts tau_value [pmc_value tau_pme_value tau_ci_lower_bound tau_ci_upper_bound] other_attributes
581 Fields are separated by the tab character. 'other_attributes' are all
582 other attributes after attribute 'transcript_id' field in the GTF
583 file. If no other attributes are given or no GTF file is provided in
584 'rsem-prepare-reference', there will be no tab after the
587 =item B<sample_name.bam, sample_name.sorted.bam and sample_name.sorted.bam.bai>
589 Only generated when --out-bam is specified.
591 'sample_name.bam' is a BAM-formatted file of read alignments in
592 genomic coordinates. Alignments of reads that have identical genomic
593 coordinates (i.e., alignments to different isoforms that share the
594 same genomic region) are collapsed into one alignment. The MAPQ field
595 of each alignment is set to max(100, floor(-10 * log10(1.0 - w) +
596 0.5)), where w is the posterior probability of that alignment being
597 the true mapping of a read. In addition, RSEM pads a new tag
598 ZW:f:value, where value is a single precision floating number
599 representing the posterior probability.
601 'sample_name.sorted.bam' and 'sample_name.sorted.bam.bai' are the
602 sorted BAM file and indices generated by samtools (included in RSEM package).
604 =item B<sample_name.stat>
606 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.
612 Assume the path to the bowtie executables is in the user's PATH environment variable. Reference files are under '/ref' with name 'mm9'.
614 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:
616 rsem-calculate-expression --phred64-quals \
623 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:
625 rsem-calculate-expression -p 8 \
630 mmliver_paired_end_quals
632 3) '/data/mmliver.fa', single-end reads without quality scores. We want to use 8 threads and generate a BAM file:
634 rsem-calculate-expression -p 8 \
638 mmliver_single_without_quals
640 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.
642 rsem-calculate-expression --bowtie-path /sw/bowtie \
644 --fragment-length-mean 150.0 \
645 --fragment-length-sd 35.0 \
654 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:
656 rsem-calculate-expression --paired-end \
659 /data/mmliver_paired_end_quals.bam \
661 mmliver_paired_end_quals