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 $strand_specific = 0;
59 my ($time_start, $time_end, $time_alignment, $time_rsem, $time_ci) = (0, 0, 0, 0, 0);
61 GetOptions("keep-intermediate-files" => \$keep_intermediate_files,
62 "no-qualities" => \$no_qual,
63 "paired-end" => \$paired_end,
64 "strand-specific" => \$strand_specific,
67 "sam-header-info=s" => \$fn_list,
69 "seed-length=i" => \$L,
70 "bowtie-path=s" => \$bowtie_path,
73 "bowtie-m=i" => \$maxHits,
74 "phred33-quals" => \$phred33,
75 "phred64-quals" => \$phred64, #solexa1.3-quals" => \$phred64,
76 "solexa-quals" => \$solexa,
77 "forward-prob=f" => \$probF,
78 "fragment-length-min=i" => \$minL,
79 "fragment-length-max=i" => \$maxL,
80 "fragment-length-mean=f" => \$mean,
81 "fragment-length-sd=f" => \$sd,
82 "estimate-rspd" => \$estRSPD,
83 "num-rspd-bins=i" => \$B,
84 "p|num-threads=i" => \$nThreads,
85 "out-bam" => \$genBamF,
86 "calc-ci" => \$calcCI,
87 "ci-memory=i" => \$NMB,
90 "h|help" => \$help) or pod2usage(-exitval => 2, -verbose => 2);
92 pod2usage(-verbose => 2) if ($help == 1);
95 #check parameters and options
97 if ($is_sam || $is_bam) {
98 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (scalar(@ARGV) != 3);
99 pod2usage(-msg => "--sam and --bam cannot be active at the same time!", -exitval => 2, -verbose => 2) if ($is_sam == 1&& $is_bam == 1);
100 pod2usage(-msg => "--bowtie-path, --bowtie-n, --bowtie-e, --bowtie-m, --phred33-quals, --phred64-quals or --solexa-quals cannot be set if input is SAM/BAM format!", -exitval => 2, -verbose => 2) if ($bowtie_path ne "" || $C != 2 || $E != 99999999 || $maxHits != 200 || $phred33 || $phred64 || $solexa);
103 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (!$paired_end && scalar(@ARGV) != 3 || $paired_end && scalar(@ARGV) != 4);
104 pod2usage(-msg => "Only one of --phred33-quals --phred64-quals/--solexa1.3-quals --solexa-suqls can be active!", -exitval => 2, -verbose => 2) if ($phred33 + $phred64 + $solexa > 1);
105 podwusage(-msg => "--sam , --bam or --sam-header-info cannot be set if use bowtie aligner to produce alignments!", -exitval => 2, -verbose => 2) if ($is_sam || $is_bam || $fn_list ne "");
108 pod2usage(-msg => "Forward probability should be in [0, 1]!", -exitval => 2, -verbose => 2) if ($probF < 0 || $probF > 1);
109 pod2usage(-msg => "Min fragment length should be at least 1!", -exitval => 2, -verbose => 2) if ($minL < 1);
110 pod2usage(-msg => "Min fragment length should be smaller or equal to max fragment length!", -exitval => 2, -verbose => 2) if ($minL > $maxL);
111 pod2usage(-msg => "The memory allocated for calculating credibility intervals should be at least 1 MB!\n", -exitval => 2, -verbose => 2) if ($NMB < 1);
112 pod2usage(-msg => "Number of threads should be at least 1!\n", -exitval => 2, -verbose => 2) if ($nThreads < 1);
113 pod2usage(-msg => "Seed length should be at least 25!\n", -exitval => 2, -verbose => 2) if ($L < 25);
115 if ($strand_specific) { $probF = 1.0; }
121 my ($refName, $sampleName, $sampleToken, $temp_dir, $stat_dir, $imdName) = ();
125 if ($no_qual) { $read_type = 2; }
126 else { $read_type = 3; }
129 if ($no_qual) { $read_type = 0; }
130 else { $read_type = 1; }
133 if (scalar(@ARGV) == 3) {
134 if ($is_sam || $is_bam) { $inpF = $ARGV[0]; }
135 else {$mate1_list = $ARGV[0]; }
137 $sampleName = $ARGV[2];
140 $mate1_list = $ARGV[0];
141 $mate2_list = $ARGV[1];
143 $sampleName = $ARGV[3];
146 my $pos = rindex($sampleName, '/');
147 if ($pos < 0) { $sampleToken = $sampleName; }
148 else { $sampleToken = substr($sampleName, $pos + 1); }
150 $temp_dir = "$sampleName.temp";
151 $stat_dir = "$sampleName.stat";
153 if (!(-d $temp_dir) && !mkdir($temp_dir)) { print "Fail to create folder $temp_dir.\n"; exit(-1); }
154 if (!(-d $stat_dir) && !mkdir($stat_dir)) { print "Fail to create folder $stat_dir.\n"; exit(-1); }
156 $imdName = "$temp_dir/$sampleToken";
158 if (!$is_sam && !$is_bam && $phred33 + $phred64 + $solexa == 0) { $phred33 = 1; }
160 my ($mate_minL, $mate_maxL) = (1, $maxL);
162 if ($bowtie_path ne "") { $bowtie_path .= "/"; }
164 my ($fn, $dir, $suf) = fileparse($0);
167 if (!$is_sam && !$is_bam) {
168 $command = $bowtie_path."bowtie";
169 if ($read_type == 0 || $read_type == 2) { $command .= " -f"; }
170 else { $command .= " -q"; }
172 if ($phred33) { $command .= " --phred33-quals"; }
173 elsif ($phred64) { $command .= " --phred64-quals"; }
174 elsif ($solexa) { $command .= " --solexa-quals"; }
175 else { print "Oh, no!!!"; exit(2); }
177 $command .= " -n $C -e $E -l $L";
179 if ($read_type == 2 || $read_type == 3) { $command .= " -I $minL -X $maxL"; }
181 if ($strand_specific || $probF == 1.0) { $command .= " --norc"; }
182 elsif ($probF == 0.0) { $command .= " --nofw"; }
184 $command .= " -p $nThreads -a -m $maxHits -S";
185 if ($quiet) { $command .= " --quiet"; }
187 $command .= " $refName";
188 if ($read_type == 0 || $read_type == 1) {
189 $command .= " $mate1_list";
192 $command .= " -1 $mate1_list -2 $mate2_list";
195 $command .= " | gzip > $imdName.sam.gz";
198 if ($mTime) { $time_start = time(); }
200 $status = system($command);
202 if ($mTime) { $time_end = time(); $time_alignment = $time_end - $time_start; }
205 print "bowtie failed! Please check if you provide correct parameters/options for the pipeline!\n";
210 $inpF = "$imdName.sam.gz";
211 $is_sam = 1; # output of bowtie is a sam file
214 if ($mTime) { $time_start = time(); }
216 $command = $dir."rsem-parse-alignments $refName $sampleName $sampleToken";
219 if ($is_sam) { $samInpType = "s"; }
220 elsif ($is_bam) { $samInpType = "b"; }
222 $command .= " $samInpType $inpF -t $read_type";
223 if ($fn_list ne "") { $command .= " -l $fn_list"; }
224 if ($tagName ne "") { $command .= " -tag $tagName"; }
225 if ($quiet) { $command .= " -q"; }
228 $status = system($command);
230 print "rsem-parse-alignments failed! Please check if you provide correct parameters/options for the pipeline!\n";
235 $command = $dir."rsem-build-read-index $gap";
237 case 0 { $command .= " 0 $quiet $imdName\_alignable.fa"; }
238 case 1 { $command .= " 1 $quiet $imdName\_alignable.fq"; }
239 case 2 { $command .= " 0 $quiet $imdName\_alignable_1.fa $imdName\_alignable_2.fa"; }
240 case 3 { $command .= " 1 $quiet $imdName\_alignable_1.fq $imdName\_alignable_2.fq"; }
243 $status = system($command);
245 print "rsem-build-read-index failed! Please check if you provide correct parameters/options for the pipeline!\n";
250 my $doesOpen = open(OUTPUT, ">$imdName.mparams");
251 if ($doesOpen == 0) { print "Cannot generate $imdName.mparams!\n"; exit(-1); }
252 print OUTPUT "$minL $maxL\n";
253 print OUTPUT "$probF\n";
254 print OUTPUT "$estRSPD\n";
256 print OUTPUT "$mate_minL $mate_maxL\n";
257 print OUTPUT "$mean $sd\n";
261 $command = $dir."rsem-run-em $refName $read_type $sampleName $sampleToken -p $nThreads";
263 $command .= " -b $samInpType $inpF";
264 if ($fn_list ne "") { $command .= " 1 $fn_list"; }
265 else { $command .= " 0"; }
267 if ($calcCI) { $command .= " --gibbs-out"; }
268 if ($quiet) { $command .= " -q"; }
271 $status = system($command);
273 print "rsem-run-em failed! Please check if you provide correct parameters/options for the pipeline!\n";
279 $command = $dir."sam/samtools sort $sampleName.bam $sampleName.sorted";
281 $status = system($command);
283 print "sam/samtools sort failed! Please check if you provide correct parameters/options for the pipeline!\n";
287 $command = $dir."sam/samtools index $sampleName.sorted.bam";
289 $status = system($command);
291 print "sam/samtools index failed! Please check if you provide correct parameters/options for the pipeline!\n";
297 &collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
298 &collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
300 if ($mTime) { $time_end = time(); $time_rsem = $time_end - $time_start; }
302 if ($mTime) { $time_start = time(); }
305 $command = $dir."rsem-run-gibbs $refName $sampleName $sampleToken $BURNIN $CHAINLEN $SAMPLEGAP";
306 # $command .= " -p $nThreads";
307 if ($quiet) { $command .= " -q"; }
309 $status = system($command);
311 print "rsem-run-gibbs failed! Please check if you provide correct parameters/options for the pipeline!\n";
316 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak1");
317 system("mv $sampleName.genes.results $imdName.genes.results.bak1");
318 &collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
319 &collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
321 $command = $dir."rsem-calculate-credibility-intervals $refName $sampleName $sampleToken $CONFIDENCE $NSPC $NMB";
322 if ($quiet) { $command .= " -q"; }
324 $status = system($command);
326 print "rsem-calculate-credibility-intervals failed! Please check if you provide correct parameters/options for the pipeline!\n";
331 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak2");
332 system("mv $sampleName.genes.results $imdName.genes.results.bak2");
333 &collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
334 &collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
337 if ($mTime) { $time_end = time(); $time_ci = $time_end - $time_start; }
339 if ($mTime) { $time_start = time(); }
341 if (!$keep_intermediate_files) {
342 $status = system("rm -rf $temp_dir");
344 print "Fail to delete the temporary folder!\n";
349 if ($mTime) { $time_end = time(); }
352 open(OUTPUT, ">$sampleName.time");
353 print OUTPUT "Alignment: $time_alignment s.\n";
354 print OUTPUT "RSEM: $time_rsem s.\n";
355 print OUTPUT "CI: $time_ci s.\n";
356 my $time_del = $time_end - $time_start;
357 print OUTPUT "Delete: $time_del s.\n";
365 my (@results, @comment) = ();
372 $local_status = open(INPUT, $inpF);
373 if ($local_status == 0) { print "Fail to open file $inpF!\n"; exit(-1); }
378 while ($line = <INPUT>) {
381 my @local_arr = split(/\t/, $line);
382 if ($cnt == 4) { @comment = @local_arr; }
383 else { push(@results, \@local_arr); }
386 push(@results, \@comment);
389 $local_status = open(OUTPUT, ">$outF");
390 if ($local_status == 0) { print "Fail to create file $outF!\n"; exit(-1); }
392 my $n = scalar(@results);
393 my $m = scalar(@{$results[0]});
394 for (my $i = 0; $i < $m; $i++) {
396 for (my $j = 0; $j < $n; $j++) { push(@out_arr, $results[$j][$i]); }
398 print OUTPUT "@out_arr\n";
408 rsem-calculate-expression
414 rsem-calculate-expression [options] upstream_read_file(s) reference_name sample_name
415 rsem-calculate-expression [options] --paired-end upstream_read_file(s) downstream_read_file(s) reference_name sample_name
416 rsem-calculate-expression [options] --sam/--bam [--paired-end] input reference_name sample_name
424 =item B<upstream_read_files(s)>
426 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.
428 =item B<downstream_read_file(s)>
430 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.
434 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.
436 =item B<reference_name>
438 The name of the reference used. The user must have run 'rsem-prepare-reference' with this reference_name before running this program.
442 The name of the sample analyzed. All output files are prefixed by this name (e.g., sample_name.genes.results)
450 =item B<--paired-end>
452 Input reads are paired-end reads. (Default: off)
454 =item B<--no-qualities>
456 Input reads do not contain quality scores. (Default: off)
458 =item B<--strand-specific>
460 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)
464 Input file is in SAM format. (Default: off)
468 Input file is in BAM format. (Default: off)
470 =item B<--sam-header-info> <file>
472 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: "")
474 =item B<-p/--num-threads> <int>
476 Number of threads to use. Both Bowtie and expression estimation will use this many threads. (Default: 1)
480 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)
484 Calculate 95% credibility intervals and posterior mean estimates. (Default: off)
486 =item B<--seed-length> <int>
488 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)
490 =item B<--tag> <string>
492 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: "")
494 =item B<--bowtie-path> <path>
496 The path to the bowtie executables. (Default: the path to the bowtie executables is assumed to be in the user's PATH environment variable)
498 =item B<--bowtie-n> <int>
500 (Bowtie parameter) max # of mismatches in the seed. (Range: 0-3, Default: 2)
502 =item B<--bowtie-e> <int>
504 (Bowtie parameter) max sum of mismatch quality scores across the alignment. (Default: 99999999)
506 =item B<--bowtie-m> <int>
508 (Bowtie parameter) suppress all alignments for a read if > <int> valid alignments exist. (Default: 200)
510 =item B<--phred33-quals>
512 Input quality scores are encoded as Phred+33. (Default: on)
514 =item B<--phred64-quals>
516 Input quality scores are encoded as Phred+64 (default for GA Pipeline ver. >= 1.3). (Default: off)
518 =item B<--solexa-quals>
520 Input quality scores are solexa encoded (from GA Pipeline ver. < 1.3). (Default: off)
522 =item B<--forward-prob> <double>
524 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)
526 =item B<--fragment-length-min> <int>
528 Minimum read/insert length allowed. This is also the value for the bowtie -I option. (Default: 1)
530 =item B<--fragment-length-max> <int>
532 Maximum read/insert length allowed. This is also the value for the bowtie -X option. (Default: 1000)
534 =item B<--fragment-length-mean> <double>
536 (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)
538 =item B<--fragment-length-sd> <double>
540 (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>)
542 =item B<--estimate-rspd>
544 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)
546 =item B<--num-rspd-bins> <int>
548 Number of bins in the RSPD. Only relevant when '--estimate-rspd' is specified. Use of the default setting is recommended. (Default: 20)
550 =item B<--ci-memory> <int>
552 Amount of memory (in MB) RSEM is allowed to use for computing credibility intervals. (Default: 1024)
554 =item B<--keep-intermediate-files>
556 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)
560 Suppress the output of logging information. (Default: off)
564 Show help information.
570 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.
572 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:
574 sort -k 1,1 -s input.sam > input.sorted.sam
576 The SAM/BAM format RSEM uses is v1.3. However, it is compatible with old SAM/BAM format.
578 The user must run 'rsem-prepare-reference' with the appropriate reference before using this program.
580 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.
582 Please note that some of the default values for the Bowtie parameters are not the same as those defined for Bowtie itself.
584 The temporary directory and all intermediate files will be removed when RSEM finishes unless '--keep-intermediate-files' is specified.
586 With the "--calc-ci" option, 95% credibility intervals and posterior mean estimates will be calculated in addition to maximum likelihood estimates.
592 =item B<sample_name.genes.results>
594 File containing gene level expression estimates. The format of each
595 line in this file is:
597 gene_id expected_counts tau_value [pmc_value tau_pme_value tau_ci_lower_bound tau_ci_upper_bound] transcript_id_list
599 Fields are separated by the tab character. Fields within "[]" are only
600 presented if '--calc-ci' is set. pme stands for posterior mean
601 estimation. pmc stands for posterior mean counts. ci_lower_bound(l)
602 means the lower bound of the credibility intervals, ci_upper_bound(u)
603 means the upper bound of the credibility intervals. So the credibility
604 interval is [l, u]. 'transcript_id_list' is a space-separated list of
605 transcript_ids belonging to the gene.
607 =item B<sample_name.isoforms.results>
609 File containing isoform level expression values. The format of each
610 line in this file is:
612 transcript_id expected_counts tau_value [pmc_value tau_pme_value tau_ci_lower_bound tau_ci_upper_bound] other_attributes
614 Fields are separated by the tab character. 'other_attributes' are all
615 other attributes after attribute 'transcript_id' field in the GTF
616 file. If no other attributes are given or no GTF file is provided in
617 'rsem-prepare-reference', there will be no tab after the
620 =item B<sample_name.bam, sample_name.sorted.bam and sample_name.sorted.bam.bai>
622 Only generated when --out-bam is specified.
624 'sample_name.bam' is a BAM-formatted file of read alignments in
625 genomic coordinates. Alignments of reads that have identical genomic
626 coordinates (i.e., alignments to different isoforms that share the
627 same genomic region) are collapsed into one alignment. The MAPQ field
628 of each alignment is set to min(100, floor(-10 * log10(1.0 - w) +
629 0.5)), where w is the posterior probability of that alignment being
630 the true mapping of a read. In addition, RSEM pads a new tag
631 ZW:f:value, where value is a single precision floating number
632 representing the posterior probability.
634 'sample_name.sorted.bam' and 'sample_name.sorted.bam.bai' are the
635 sorted BAM file and indices generated by samtools (included in RSEM package).
637 =item B<sample_name.stat>
639 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.
645 Assume the path to the bowtie executables is in the user's PATH environment variable. Reference files are under '/ref' with name 'mm9'.
647 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:
649 rsem-calculate-expression --phred64-quals \
656 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:
658 rsem-calculate-expression -p 8 \
663 mmliver_paired_end_quals
665 3) '/data/mmliver.fa', single-end reads without quality scores. We want to use 8 threads and generate a BAM file:
667 rsem-calculate-expression -p 8 \
671 mmliver_single_without_quals
673 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.
675 rsem-calculate-expression --bowtie-path /sw/bowtie \
677 --fragment-length-mean 150.0 \
678 --fragment-length-sd 35.0 \
687 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:
689 rsem-calculate-expression --paired-end \
692 /data/mmliver_paired_end_quals.bam \
694 mmliver_paired_end_quals