7 use lib $FindBin::RealBin;
8 use rsem_perl_utils qw(runCommand collectResults showVersionInfo);
11 @PATH = ($FindBin::RealBin, "$FindBin::RealBin/sam", @PATH);
19 my $CONFIDENCE = 0.95;
22 my $NMB = 1024; # default
23 my $SortMem = "1G"; # default as 1G per thread
27 my $read_type = 1; # default, single end with qual
34 my $chunkMbs = 0; # 0 = use bowtie default
55 my $genBamF = 1; # default is generating transcript bam file
56 my $genGenomeBamF = 0;
65 my $keep_intermediate_files = 0;
67 my $strand_specific = 0;
70 my $bowtie2_path = "";
71 my $bowtie2_mismatch_rate = 0.1;
73 my $bowtie2_sensitivity_level = "sensitive"; # must be one of "very_fast", "fast", "sensitive", "very_sensitive"
80 my ($time_start, $time_end, $time_alignment, $time_rsem, $time_ci) = (0, 0, 0, 0, 0);
86 my ($refName, $sampleName, $sampleToken, $temp_dir, $stat_dir, $imdName, $statName) = ();
91 GetOptions("keep-intermediate-files" => \$keep_intermediate_files,
92 "temporary-folder=s" => \$temp_dir,
93 "no-qualities" => \$no_qual,
94 "paired-end" => \$paired_end,
95 "strand-specific" => \$strand_specific,
98 "sam-header-info=s" => \$fn_list,
100 "seed-length=i" => \$L,
101 "bowtie-path=s" => \$bowtie_path,
104 "bowtie-m=i" => \$maxHits,
105 "bowtie-chunkmbs=i" => \$chunkMbs,
106 "phred33-quals" => \$phred33,
107 "phred64-quals" => \$phred64, #solexa1.3-quals" => \$phred64,
108 "solexa-quals" => \$solexa,
109 "bowtie2" => \$bowtie2,
110 "bowtie2-path=s" => \$bowtie2_path,
111 "bowtie2-mismatch-rate=f" => \$bowtie2_mismatch_rate,
112 "bowtie2-k=i" => \$bowtie2_k,
113 "bowtie2-sensitivity-level=s" => \$bowtie2_sensitivity_level,
114 "forward-prob=f" => \$probF,
115 "fragment-length-min=i" => \$minL,
116 "fragment-length-max=i" => \$maxL,
117 "fragment-length-mean=f" => \$mean,
118 "fragment-length-sd=f" => \$sd,
119 "estimate-rspd" => \$estRSPD,
120 "num-rspd-bins=i" => \$B,
121 "p|num-threads=i" => \$nThreads,
122 "no-bam-output" => sub { $genBamF = 0; },
123 "output-genome-bam" => \$genGenomeBamF,
124 "sampling-for-bam" => \$sampling,
125 "calc-pme" => \$calcPME,
126 "calc-ci" => \$calcCI,
127 "ci-memory=i" => \$NMB,
128 "samtools-sort-mem=s" => \$SortMem,
131 "version" => \$version,
132 "q|quiet" => \$quiet,
133 "h|help" => \$help) or pod2usage(-exitval => 2, -verbose => 2);
135 pod2usage(-verbose => 2) if ($help == 1);
136 &showVersionInfo($FindBin::RealBin) if ($version == 1);
138 #check parameters and options
140 if ($is_sam || $is_bam) {
141 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (scalar(@ARGV) != 3);
142 pod2usage(-msg => "--sam and --bam cannot be active at the same time!", -exitval => 2, -verbose => 2) if ($is_sam == 1&& $is_bam == 1);
143 pod2usage(-msg => "--bowtie-path, --bowtie-n, --bowtie-e, --bowtie-m, --phred33-quals, --phred64-quals, --solexa-quals, --bowtie2, --bowtie2-path, --bowtie2-mismatch-rate, --bowtie2-k and --bowtie2-sensitivity-level cannot be set if input is SAM/BAM format!", -exitval => 2, -verbose => 2) if ($bowtie_path ne "" || $C != 2 || $E != 99999999 || $maxHits != 200 || $phred33 || $phred64 || $solexa || $bowtie2 || $bowtie2_path ne "" || $bowtie2_mismatch_rate != 0.1 || $bowtie2_k != 200 || $bowtie2_sensitivity_level ne "sensitive");
146 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (!$paired_end && scalar(@ARGV) != 3 || $paired_end && scalar(@ARGV) != 4);
147 pod2usage(-msg => "If --no-qualities is set, neither --phred33-quals, --phred64-quals or --solexa-quals can be active!", -exitval => 2, -verbose => 2) if ($no_qual && ($phred33 + $phred64 + $solexa > 0));
148 pod2usage(-msg => "Only one of --phred33-quals, --phred64-quals, and --solexa-quals can be active!", -exitval => 2, -verbose => 2) if ($phred33 + $phred64 + $solexa > 1);
149 pod2usage(-msg => "--sam , --bam or --sam-header-info cannot be set if use bowtie/bowtie2 aligner to produce alignments!", -exitval => 2, -verbose => 2) if ($is_sam || $is_bam || $fn_list ne "");
150 pod2usage(-msg => "--bowtie2-path, --bowtie2-mismatch-rate, --bowtie2-k and --bowtie2-sensitivity-level cannot be set if bowtie aligner is used!", -exitval => 2, -verbose => 2) if (!$bowtie2 && ($bowtie2_path ne "" || $bowtie2_mismatch_rate != 0.1 || $bowtie2_k != 200 || $bowtie2_sensitivity_level ne "sensitive"));
151 pod2usage(-msg => "--bowtie-path, --bowtie-n, --bowtie-e, --bowtie-m cannot be set if bowtie2 aligner is used!", -exitval => 2, -verbose => 2) if ($bowtie2 && ($bowtie_path ne "" || $C != 2 || $E != 99999999 || $maxHits != 200));
152 pod2usage(-msg => "Mismatch rate must be within [0, 1]!", -exitval => 2, -verbose => 2) if ($bowtie2 && ($bowtie2_mismatch_rate < 0.0 || $bowtie2_mismatch_rate > 1.0));
153 pod2usage(-msg => "Sensitivity level must be one of \"very_fast\", \"fast\", \"sensitive\", and \"very_sensitive\"!", -exitval => 2, -verbose => 2) if ($bowtie2 && (($bowtie2_sensitivity_level ne "very_fast") && ($bowtie2_sensitivity_level ne "fast") && ($bowtie2_sensitivity_level ne "sensitive") && ($bowtie2_sensitivity_level ne "very_sensitive")));
156 pod2usage(-msg => "Forward probability should be in [0, 1]!", -exitval => 2, -verbose => 2) if ($probF < 0 || $probF > 1);
157 pod2usage(-msg => "Min fragment length should be at least 1!", -exitval => 2, -verbose => 2) if ($minL < 1);
158 pod2usage(-msg => "Min fragment length should be smaller or equal to max fragment length!", -exitval => 2, -verbose => 2) if ($minL > $maxL);
159 pod2usage(-msg => "The memory allocated for calculating credibility intervals should be at least 1 MB!\n", -exitval => 2, -verbose => 2) if ($NMB < 1);
160 pod2usage(-msg => "Number of threads should be at least 1!\n", -exitval => 2, -verbose => 2) if ($nThreads < 1);
161 pod2usage(-msg => "Seed length should be at least 5!\n", -exitval => 2, -verbose => 2) if ($L < 5);
162 pod2usage(-msg => "--sampling-for-bam cannot be specified if --no-bam-output is specified!\n", -exitval => 2, -verbose => 2) if ($sampling && !$genBamF);
163 pod2usage(-msg => "--output-genome-bam cannot be specified if --no-bam-output is specified!\n", -exitval => 2, -verbose => 2) if ($genGenomeBamF && !$genBamF);
164 pod2usage(-msg => "The seed for random number generator must be a non-negative 32bit integer!\n", -exitval => 2, -verbose => 2) if (($seed ne "NULL") && ($seed < 0 || $seed > 0xffffffff));
166 if ($L < 25) { print "Warning: the seed length set is less than 25! This is only allowed if the references are not added poly(A) tails.\n"; }
168 if ($strand_specific) { $probF = 1.0; }
171 if ($no_qual) { $read_type = 2; }
172 else { $read_type = 3; }
175 if ($no_qual) { $read_type = 0; }
176 else { $read_type = 1; }
179 if (scalar(@ARGV) == 3) {
180 if ($is_sam || $is_bam) { $inpF = $ARGV[0]; }
181 else {$mate1_list = $ARGV[0]; }
183 $sampleName = $ARGV[2];
186 $mate1_list = $ARGV[0];
187 $mate2_list = $ARGV[1];
189 $sampleName = $ARGV[3];
192 if (((-e "$refName.ta") && !(-e "$refName.gt")) || (!(-e "$refName.ta") && (-e "$refName.gt"))) {
193 print "Allele-specific expression related reference files are corrupted!\n";
197 $alleleS = (-e "$refName.ta") && (-e "$refName.gt");
199 if ($genGenomeBamF) {
200 open(INPUT, "$refName.ti");
201 my $line = <INPUT>; chomp($line);
203 my ($M, $type) = split(/ /, $line);
204 pod2usage(-msg => "No genome information provided, so genome bam file cannot be generated!\n", -exitval => 2, -verbose => 2) if ($type != 0);
207 my $pos = rindex($sampleName, '/');
208 if ($pos < 0) { $sampleToken = $sampleName; }
209 else { $sampleToken = substr($sampleName, $pos + 1); }
211 if ($temp_dir eq "") { $temp_dir = "$sampleName.temp"; }
212 $stat_dir = "$sampleName.stat";
214 if (!(-d $temp_dir) && !mkdir($temp_dir)) { print "Fail to create folder $temp_dir.\n"; exit(-1); }
215 if (!(-d $stat_dir) && !mkdir($stat_dir)) { print "Fail to create folder $stat_dir.\n"; exit(-1); }
217 $imdName = "$temp_dir/$sampleToken";
218 $statName = "$stat_dir/$sampleToken";
220 if (!$is_sam && !$is_bam && !$no_qual && ($phred33 + $phred64 + $solexa == 0)) { $phred33 = 1; }
222 my ($mate_minL, $mate_maxL) = (1, $maxL);
224 if ($bowtie_path ne "") { $bowtie_path .= "/"; }
225 if ($bowtie2_path ne "") { $bowtie2_path .= "/"; }
229 if (!$is_sam && !$is_bam) {
231 $command = $bowtie_path."bowtie";
232 if ($no_qual) { $command .= " -f"; }
233 else { $command .= " -q"; }
235 if ($phred33) { $command .= " --phred33-quals"; }
236 elsif ($phred64) { $command .= " --phred64-quals"; }
237 elsif ($solexa) { $command .= " --solexa-quals"; }
239 $command .= " -n $C -e $E -l $L";
240 if ($read_type == 2 || $read_type == 3) { $command .= " -I $minL -X $maxL"; }
241 if ($chunkMbs > 0) { $command .= " --chunkmbs $chunkMbs"; }
243 if ($strand_specific || $probF == 1.0) { $command .= " --norc"; }
244 elsif ($probF == 0.0) { $command .= " --nofw"; }
246 $command .= " -p $nThreads -a -m $maxHits -S";
247 if ($quiet) { $command .= " --quiet"; }
249 $command .= " $refName";
250 if ($read_type == 0 || $read_type == 1) {
251 $command .= " $mate1_list";
254 $command .= " -1 $mate1_list -2 $mate2_list";
257 # pipe to samtools to generate a BAM file
258 $command .= " | samtools view -S -b -o $imdName.bam -";
261 $command = $bowtie2_path."bowtie2";
262 if ($no_qual) { $command .= " -f"; }
263 else { $command .= " -q"; }
265 if ($phred33) { $command .= " --phred33"; }
266 elsif ($phred64) { $command .= " --phred64"; }
267 elsif ($solexa) { $command .= " --solexa-quals"; }
269 if ($bowtie2_sensitivity_level eq "very_fast") { $command .= " --very-fast"; }
270 elsif ($bowtie2_sensitivity_level eq "fast") { $command .= " --fast"; }
271 elsif ($bowtie2_sensitivity_level eq "sensitive") { $command .= " --sensitive"; }
272 else { $command .= " --very-sensitive"; }
274 $command .= " --dpad 0 --gbar 99999999 --mp 1,1 --np 1 --score-min L,0,-$bowtie2_mismatch_rate";
276 if ($read_type == 2 || $read_type == 3) { $command .= " -I $minL -X $maxL --no-mixed --no-discordant"; }
278 if ($strand_specific || $probF == 1.0) { $command .= " --norc"; }
279 elsif ($probF == 0.0) { $command .= " --nofw"; }
281 $command .= " -p $nThreads -k $bowtie2_k";
282 if ($quiet) { $command .= " --quiet"; }
284 $command .= " -x $refName";
285 if ($read_type == 0 || $read_type == 1) {
286 $command .= " -U $mate1_list";
289 $command .= " -1 $mate1_list -2 $mate2_list";
292 # pipe to samtools to generate a BAM file
293 $command .= " | samtools view -S -b -o $imdName.bam -";
296 if ($mTime) { $time_start = time(); }
298 &runCommand($command);
300 if ($mTime) { $time_end = time(); $time_alignment = $time_end - $time_start; }
302 $inpF = "$imdName.bam";
303 $is_bam = 1; # alignments are outputed as a BAM file
306 if ($mTime) { $time_start = time(); }
308 $command = "rsem-parse-alignments $refName $imdName $statName";
311 if ($is_sam) { $samInpType = "s"; }
312 elsif ($is_bam) { $samInpType = "b"; }
314 $command .= " $samInpType $inpF -t $read_type";
315 if ($fn_list ne "") { $command .= " -l $fn_list"; }
316 if ($tagName ne "") { $command .= " -tag $tagName"; }
317 if ($quiet) { $command .= " -q"; }
319 &runCommand($command);
321 $command = "rsem-build-read-index $gap";
322 if ($read_type == 0) { $command .= " 0 $quiet $imdName\_alignable.fa"; }
323 elsif ($read_type == 1) { $command .= " 1 $quiet $imdName\_alignable.fq"; }
324 elsif ($read_type == 2) { $command .= " 0 $quiet $imdName\_alignable_1.fa $imdName\_alignable_2.fa"; }
325 elsif ($read_type == 3) { $command .= " 1 $quiet $imdName\_alignable_1.fq $imdName\_alignable_2.fq"; }
326 else { print "Impossible! read_type is not in [1,2,3,4]!\n"; exit(-1); }
327 &runCommand($command);
329 my $doesOpen = open(OUTPUT, ">$imdName.mparams");
330 if ($doesOpen == 0) { print "Cannot generate $imdName.mparams!\n"; exit(-1); }
331 print OUTPUT "$minL $maxL\n";
332 print OUTPUT "$probF\n";
333 print OUTPUT "$estRSPD\n";
335 print OUTPUT "$mate_minL $mate_maxL\n";
336 print OUTPUT "$mean $sd\n";
341 if ($seed ne "NULL") {
343 for (my $i = 0; $i < 3; $i++) {
344 push(@seeds, int(rand(1 << 32)));
348 $command = "rsem-run-em $refName $read_type $sampleName $imdName $statName -p $nThreads";
350 $command .= " -b $samInpType $inpF";
351 if ($fn_list ne "") { $command .= " 1 $fn_list"; }
352 else { $command .= " 0"; }
353 if ($sampling) { $command .= " --sampling"; }
354 if ($seed ne "NULL") { $command .= " --seed $seeds[0]"; }
356 if ($calcPME || $calcCI) { $command .= " --gibbs-out"; }
357 if ($quiet) { $command .= " -q"; }
359 &runCommand($command);
362 &collectResults("allele", "$imdName.allele_res", "$sampleName.alleles.results"); # allele level
363 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
364 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
367 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
368 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
372 $command = "samtools sort -@ $nThreads -m $SortMem $sampleName.transcript.bam $sampleName.transcript.sorted";
373 &runCommand($command);
374 $command = "samtools index $sampleName.transcript.sorted.bam";
375 &runCommand($command);
377 if ($genGenomeBamF) {
378 $command = "rsem-tbam2gbam $refName $sampleName.transcript.bam $sampleName.genome.bam";
379 &runCommand($command);
380 $command = "samtools sort -@ $nThreads -m $SortMem $sampleName.genome.bam $sampleName.genome.sorted";
381 &runCommand($command);
382 $command = "samtools index $sampleName.genome.sorted.bam";
383 &runCommand($command);
387 if ($mTime) { $time_end = time(); $time_rsem = $time_end - $time_start; }
389 if ($mTime) { $time_start = time(); }
391 if ($calcPME || $calcCI ) {
392 $command = "rsem-run-gibbs $refName $imdName $statName $BURNIN $NCV $SAMPLEGAP";
393 $command .= " -p $nThreads";
394 if ($seed ne "NULL") { $command .= " --seed $seeds[1]"; }
395 if ($quiet) { $command .= " -q"; }
396 &runCommand($command);
399 if ($calcPME || $calcCI) {
401 system("mv $sampleName.alleles.results $imdName.alleles.results.bak1");
402 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak1");
403 system("mv $sampleName.genes.results $imdName.genes.results.bak1");
404 &collectResults("allele", "$imdName.allele_res", "$sampleName.alleles.results"); # allele level
405 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
406 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
409 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak1");
410 system("mv $sampleName.genes.results $imdName.genes.results.bak1");
411 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
412 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
417 $command = "rsem-calculate-credibility-intervals $refName $imdName $statName $CONFIDENCE $NCV $NSPC $NMB";
418 $command .= " -p $nThreads";
419 if ($seed ne "NULL") { $command .= " --seed $seeds[2]"; }
420 if ($quiet) { $command .= " -q"; }
421 &runCommand($command);
424 system("mv $sampleName.alleles.results $imdName.alleles.results.bak2");
425 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak2");
426 system("mv $sampleName.genes.results $imdName.genes.results.bak2");
427 &collectResults("allele", "$imdName.allele_res", "$sampleName.alleles.results"); # allele level
428 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
429 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
432 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak2");
433 system("mv $sampleName.genes.results $imdName.genes.results.bak2");
434 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
435 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
439 if ($mTime) { $time_end = time(); $time_ci = $time_end - $time_start; }
441 if ($mTime) { $time_start = time(); }
443 if (!$keep_intermediate_files) {
444 &runCommand("rm -rf $temp_dir", "Fail to delete the temporary folder!");
447 if ($mTime) { $time_end = time(); }
450 open(OUTPUT, ">$sampleName.time");
451 print OUTPUT "Aligning reads: $time_alignment s.\n";
452 print OUTPUT "Estimating expression levels: $time_rsem s.\n";
453 print OUTPUT "Calculating credibility intervals: $time_ci s.\n";
454 # my $time_del = $time_end - $time_start;
455 # print OUTPUT "Delete: $time_del s.\n";
463 rsem-calculate-expression
467 rsem-calculate-expression [options] upstream_read_file(s) reference_name sample_name
468 rsem-calculate-expression [options] --paired-end upstream_read_file(s) downstream_read_file(s) reference_name sample_name
469 rsem-calculate-expression [options] --sam/--bam [--paired-end] input reference_name sample_name
475 =item B<upstream_read_files(s)>
477 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.
479 =item B<downstream_read_file(s)>
481 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.
485 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.
487 =item B<reference_name>
489 The name of the reference used. The user must have run 'rsem-prepare-reference' with this reference_name before running this program.
493 The name of the sample analyzed. All output files are prefixed by this name (e.g., sample_name.genes.results)
501 =item B<--paired-end>
503 Input reads are paired-end reads. (Default: off)
505 =item B<--no-qualities>
507 Input reads do not contain quality scores. (Default: off)
509 =item B<--strand-specific>
511 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/Bowtie 2 aligner, the '--norc' Bowtie/Bowtie 2 option will be used, which disables alignment to the reverse strand of transcripts. (Default: off)
515 Input file is in SAM format. (Default: off)
519 Input file is in BAM format. (Default: off)
521 =item B<--sam-header-info> <file>
523 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: "")
525 =item B<-p/--num-threads> <int>
527 Number of threads to use. Both Bowtie/Bowtie2, expression estimation and 'samtools sort' will use this many threads. (Default: 1)
529 =item B<--no-bam-output>
531 Do not output any BAM file. (Default: off)
533 =item B<--output-genome-bam>
535 Generate a BAM file, 'sample_name.genome.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.genome.sorted.bam' and 'sample_name.genome.sorted.bam.bai' will be generated. (Default: off)
537 =item B<--sampling-for-bam>
539 When RSEM generates a BAM file, instead of outputing all alignments a read has with their posterior probabilities, one alignment is sampled according to the posterior probabilities. The sampling procedure includes the alignment to the "noise" transcript, which does not appear in the BAM file. Only the sampled alignment has a weight of 1. All other alignments have weight 0. If the "noise" transcript is sampled, all alignments appeared in the BAM file should have weight 0. (Default: off)
541 =item B<--seed> <uint32>
543 Set the seed for the random number generators used in calculating posterior mean estimates and credibility intervals. The seed must be a non-negative 32 bit interger. (Default: off)
547 Run RSEM's collapsed Gibbs sampler to calculate posterior mean estimates. (Default: off)
551 Calculate 95% credibility intervals and posterior mean estimates. (Default: off)
553 =item B<--seed-length> <int>
555 Seed length used by the read aligner. Providing the correct value is important for RSEM. If RSEM runs Bowtie, it uses this value for Bowtie's seed length parameter. Any read with its or at least one of its mates' (for paired-end reads) length less than this value will be ignored. If the references are not added poly(A) tails, the minimum allowed value is 5, otherwise, the minimum allowed value is 25. Note that this script will only check if the value >= 5 and give a warning message if the value < 25 but >= 5. (Default: 25)
557 =item B<--tag> <string>
559 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: "")
561 =item B<--bowtie-path> <path>
563 The path to the Bowtie executables. (Default: the path to the Bowtie executables is assumed to be in the user's PATH environment variable)
565 =item B<--bowtie-n> <int>
567 (Bowtie parameter) max # of mismatches in the seed. (Range: 0-3, Default: 2)
569 =item B<--bowtie-e> <int>
571 (Bowtie parameter) max sum of mismatch quality scores across the alignment. (Default: 99999999)
573 =item B<--bowtie-m> <int>
575 (Bowtie parameter) suppress all alignments for a read if > <int> valid alignments exist. (Default: 200)
577 =item B<--bowtie-chunkmbs> <int>
579 (Bowtie parameter) memory allocated for best first alignment calculation (Default: 0 - use Bowtie's default)
581 =item B<--phred33-quals>
583 Input quality scores are encoded as Phred+33. (Default: on)
585 =item B<--phred64-quals>
587 Input quality scores are encoded as Phred+64 (default for GA Pipeline ver. >= 1.3). (Default: off)
589 =item B<--solexa-quals>
591 Input quality scores are solexa encoded (from GA Pipeline ver. < 1.3). (Default: off)
595 Use Bowtie 2 instead of Bowtie to align reads. Since currently RSEM does not handle indel, local and discordant alignments, the Bowtie2 parameters are set in a way to avoid those alignments. In particular, we use options '--sensitive --dpad 0 --gbar 99999999 --mp 1,1 --np 1 --score-min L,0,-0.1' by default. "-0.1", the last parameter of '--score-min' is the negative value of the maximum mismatch rate allowed. This rate can be set by option '--bowtie2-mismatch-rate'. If reads are paired-end, we additionally use options '--no-mixed' and '--no-discordant'. (Default: off)
597 =item B<--bowtie2-path> <path>
599 (Bowtie 2 parameter) The path to the Bowtie 2 executables. (Default: the path to the Bowtie 2 executables is assumed to be in the user's PATH environment variable)
601 =item B<--bowtie2-mismatch-rate> <double>
603 (Bowtie 2 parameter) The maximum mismatch rate allowed. (Default: 0.1)
605 =item B<--bowtie2-k> <int>
607 (Bowtie 2 parameter) Find up to <int> alignments per read. (Default: 200)
609 =item B<--bowtie2-sensitivity-level> <string>
611 (Bowtie 2 parameter) Set Bowtie 2's preset options in --end-to-end mode. This option controls how hard Bowtie 2 tries to find alignments. <string> must be one of "very_fast", "fast", "sensitive" and "very_sensitive". The four candidates correspond to Bowtie 2's "--very-fast", "--fast", "--sensitive" and "--very-sensitive" options. (Default: "sensitive" - use Bowtie 2's default)
613 =item B<--forward-prob> <double>
615 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)
617 =item B<--fragment-length-min> <int>
619 Minimum read/insert length allowed. This is also the value for the Bowtie/Bowtie2 -I option. (Default: 1)
621 =item B<--fragment-length-max> <int>
623 Maximum read/insert length allowed. This is also the value for the Bowtie/Bowtie 2 -X option. (Default: 1000)
625 =item B<--fragment-length-mean> <double>
627 (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)
629 =item B<--fragment-length-sd> <double>
631 (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>)
633 =item B<--estimate-rspd>
635 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)
637 =item B<--num-rspd-bins> <int>
639 Number of bins in the RSPD. Only relevant when '--estimate-rspd' is specified. Use of the default setting is recommended. (Default: 20)
641 =item B<--ci-memory> <int>
643 Maximum size (in memory, MB) of the auxiliary buffer used for computing credibility intervals (CI). Set it larger for a faster CI calculation. However, leaving 2 GB memory free for other usage is recommended. (Default: 1024)
645 =item B<--samtools-sort-mem> <string>
647 Set the maximum memory per thread that can be used by 'samtools sort'. <string> represents the memory and accepts suffices 'K/M/G'. RSEM will pass <string> to the '-m' option of 'samtools sort'. Please note that the default used here is different from the default used by samtools. (Default: 1G)
649 =item B<--keep-intermediate-files>
651 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)
653 =item B<--temporary-folder> <string>
655 Set where to put the temporary files generated by RSEM. If the folder specified does not exist, RSEM will try to create it. (Default: sample_name.temp)
659 Output time consumed by each step of RSEM to 'sample_name.time'. (Default: off)
663 Suppress the output of logging information. (Default: off)
667 Show help information.
671 Show version information.
677 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 that they align against the indices generated by 'rsem-prepare-reference' and the alignment file satisfies the requirements mentioned in ARGUMENTS section.
679 One simple way to make the alignment file satisfying RSEM's requirements (assuming the aligner used put mates in a paired-end read adjacent) is to use 'convert-sam-for-rsem' script. This script only accept SAM format files as input. If a BAM format file is obtained, please use samtools to convert it to a SAM file first. For example, if '/ref/mouse_125' is the 'reference_name' and the SAM file is named 'input.sam', you can run the following command:
681 convert-sam-for-rsem /ref/mouse_125 input.sam -o input_for_rsem.sam
683 For details, please refer to 'convert-sam-for-rsem's documentation page.
685 The SAM/BAM format RSEM uses is v1.4. However, it is compatible with old SAM/BAM format. However, RSEM cannot recognize 0x100 in the FLAG field. In addition, RSEM requires SEQ and QUAL are not '*'.
687 The user must run 'rsem-prepare-reference' with the appropriate reference before using this program.
689 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.
691 Please note that some of the default values for the Bowtie parameters are not the same as those defined for Bowtie itself.
693 The temporary directory and all intermediate files will be removed when RSEM finishes unless '--keep-intermediate-files' is specified.
695 With the '--calc-ci' option, 95% credibility intervals and posterior mean estimates will be calculated in addition to maximum likelihood estimates.
701 =item B<sample_name.isoforms.results>
703 File containing isoform level expression estimates. The first line
704 contains column names separated by the tab character. The format of
705 each line in the rest of this file is:
707 transcript_id gene_id length effective_length expected_count TPM FPKM IsoPct [posterior_mean_count posterior_standard_deviation_of_count pme_TPM pme_FPKM IsoPct_from_pme_TPM TPM_ci_lower_bound TPM_ci_upper_bound FPKM_ci_lower_bound FPKM_ci_upper_bound]
709 Fields are separated by the tab character. Fields within "[]" are
710 optional. They will not be presented if neither '--calc-pme' nor
713 'transcript_id' is the transcript name of this transcript. 'gene_id'
714 is the gene name of the gene which this transcript belongs to (denote
715 this gene as its parent gene). If no gene information is provided,
716 'gene_id' and 'transcript_id' are the same.
718 'length' is this transcript's sequence length (poly(A) tail is not
719 counted). 'effective_length' counts only the positions that can
720 generate a valid fragment. If no poly(A) tail is added,
721 'effective_length' is equal to transcript length - mean fragment
722 length + 1. If one transcript's effective length is less than 1, this
723 transcript's both effective length and abundance estimates are set to
726 'expected_count' is the sum of the posterior probability of each read
727 comes from this transcript over all reads. Because 1) each read
728 aligning to this transcript has a probability of being generated from
729 background noise; 2) RSEM may filter some alignable low quality reads,
730 the sum of expected counts for all transcript are generally less than
731 the total number of reads aligned.
733 'TPM' stands for Transcripts Per Million. It is a relative measure of
734 transcript abundance. The sum of all transcripts' TPM is 1
735 million. 'FPKM' stands for Fragments Per Kilobase of transcript per
736 Million mapped reads. It is another relative measure of transcript
737 abundance. If we define l_bar be the mean transcript length in a
738 sample, which can be calculated as
740 l_bar = \sum_i TPM_i / 10^6 * effective_length_i (i goes through every transcript),
742 the following equation is hold:
744 FPKM_i = 10^3 / l_bar * TPM_i.
746 We can see that the sum of FPKM is not a constant across samples.
748 'IsoPct' stands for isoform percentage. It is the percentage of this
749 transcript's abandunce over its parent gene's abandunce. If its parent
750 gene has only one isoform or the gene information is not provided,
751 this field will be set to 100.
753 'posterior_mean_count', 'pme_TPM', 'pme_FPKM' are posterior mean
754 estimates calculated by RSEM's Gibbs
755 sampler. 'posterior_standard_deviation_of_count' is the posterior
756 standard deviation of counts. 'IsoPct_from_pme_TPM' is the isoform
757 percentage calculated from 'pme_TPM' values.
759 'TPM_ci_lower_bound', 'TPM_ci_upper_bound', 'FPKM_ci_lower_bound' and
760 'FPKM_ci_upper_bound' are lower(l) and upper(u) bounds of 95%
761 credibility intervals for TPM and FPKM values. The bounds are
762 inclusive (i.e. [l, u]).
764 =item B<sample_name.genes.results>
766 File containing gene level expression estimates. The first line
767 contains column names separated by the tab character. The format of
768 each line in the rest of this file is:
770 gene_id transcript_id(s) length effective_length expected_count TPM FPKM [posterior_mean_count posterior_standard_deviation_of_count pme_TPM pme_FPKM TPM_ci_lower_bound TPM_ci_upper_bound FPKM_ci_lower_bound FPKM_ci_upper_bound]
772 Fields are separated by the tab character. Fields within "[]" are
773 optional. They will not be presented if neither '--calc-pme' nor
776 'transcript_id(s)' is a comma-separated list of transcript_ids
777 belonging to this gene. If no gene information is provided, 'gene_id'
778 and 'transcript_id(s)' are identical (the 'transcript_id').
780 A gene's 'length' and 'effective_length' are
781 defined as the weighted average of its transcripts' lengths and
782 effective lengths (weighted by 'IsoPct'). A gene's abundance estimates
783 are just the sum of its transcripts' abundance estimates.
785 =item B<sample_name.alleles.results>
787 Only generated when the RSEM references are built with allele-specific
790 This file contains allele level expression estimates for
791 allele-specific expression calculation. The first line
792 contains column names separated by the tab character. The format of
793 each line in the rest of this file is:
795 allele_id transcript_id gene_id length effective_length expected_count TPM FPKM AlleleIsoPct AlleleGenePct [posterior_mean_count posterior_standard_deviation_of_count pme_TPM pme_FPKM AlleleIsoPct_from_pme_TPM AlleleGenePct_from_pme_TPM TPM_ci_lower_bound TPM_ci_upper_bound FPKM_ci_lower_bound FPKM_ci_upper_bound]
797 Fields are separated by the tab character. Fields within "[]" are
798 optional. They will not be presented if neither '--calc-pme' nor
801 'allele_id' is the allele-specific name of this allele-specific transcript.
803 'AlleleIsoPct' stands for allele-specific percentage on isoform
804 level. It is the percentage of this allele-specific transcript's
805 abundance over its parent transcript's abundance. If its parent
806 transcript has only one allele variant form, this field will be set to
809 'AlleleGenePct' stands for allele-specific percentage on gene
810 level. It is the percentage of this allele-specific transcript's
811 abundance over its parent gene's abundance.
813 'AlleleIsoPct_from_pme_TPM' and 'AlleleGenePct_from_pme_TPM' have
814 similar meanings. They are calculated based on posterior mean
817 Please note that if this file is present, the fields 'length' and
818 'effective_length' in 'sample_name.isoforms.results' should be
819 interpreted similarly as the corresponding definitions in
820 'sample_name.genes.results'.
822 =item B<sample_name.transcript.bam, sample_name.transcript.sorted.bam and sample_name.transcript.sorted.bam.bai>
824 Only generated when --no-bam-output is not specified.
826 'sample_name.transcript.bam' is a BAM-formatted file of read
827 alignments in transcript coordinates. The MAPQ field of each alignment
828 is set to min(100, floor(-10 * log10(1.0 - w) + 0.5)), where w is the
829 posterior probability of that alignment being the true mapping of a
830 read. In addition, RSEM pads a new tag ZW:f:value, where value is a
831 single precision floating number representing the posterior
832 probability. Because this file contains all alignment lines produced
833 by bowtie or user-specified aligners, it can also be used as a
834 replacement of the aligner generated BAM/SAM file. For paired-end
835 reads, if one mate has alignments but the other does not, this file
836 marks the alignable mate as "unmappable" (flag bit 0x4) and appends an
837 optional field "Z0:A:!".
839 'sample_name.transcript.sorted.bam' and
840 'sample_name.transcript.sorted.bam.bai' are the sorted BAM file and
841 indices generated by samtools (included in RSEM package).
843 =item B<sample_name.genome.bam, sample_name.genome.sorted.bam and sample_name.genome.sorted.bam.bai>
845 Only generated when --no-bam-output is not specified and --output-genome-bam is specified.
847 'sample_name.genome.bam' is a BAM-formatted file of read alignments in
848 genomic coordinates. Alignments of reads that have identical genomic
849 coordinates (i.e., alignments to different isoforms that share the
850 same genomic region) are collapsed into one alignment. The MAPQ field
851 of each alignment is set to min(100, floor(-10 * log10(1.0 - w) +
852 0.5)), where w is the posterior probability of that alignment being
853 the true mapping of a read. In addition, RSEM pads a new tag
854 ZW:f:value, where value is a single precision floating number
855 representing the posterior probability. If an alignment is spliced, a
856 XS:A:value tag is also added, where value is either '+' or '-'
857 indicating the strand of the transcript it aligns to.
859 'sample_name.genome.sorted.bam' and 'sample_name.genome.sorted.bam.bai' are the
860 sorted BAM file and indices generated by samtools (included in RSEM package).
862 =item B<sample_name.time>
864 Only generated when --time is specified.
866 It contains time (in seconds) consumed by aligning reads, estimating expression levels and calculating credibility intervals.
868 =item B<sample_name.stat>
870 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.
876 Assume the path to the bowtie executables is in the user's PATH environment variable. Reference files are under '/ref' with name 'mouse_125'.
878 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 genome BAM file:
880 rsem-calculate-expression --phred64-quals \
882 --output-genome-bam \
887 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 genome BAM file:
889 rsem-calculate-expression -p 8 \
894 mmliver_paired_end_quals
896 3) '/data/mmliver.fa', single-end reads without quality scores. We want to use 8 threads:
898 rsem-calculate-expression -p 8 \
902 mmliver_single_without_quals
904 4) Data are the same as 1). This time we assume the bowtie executables are under '/sw/bowtie'. 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:
906 rsem-calculate-expression --bowtie-path /sw/bowtie \
908 --fragment-length-mean 150.0 \
909 --fragment-length-sd 35.0 \
911 --output-genome-bam \
918 5) '/data/mmliver_paired_end_quals.bam', paired-end reads with quality scores. We want to use 8 threads:
920 rsem-calculate-expression --paired-end \
923 /data/mmliver_paired_end_quals.bam \
925 mmliver_paired_end_quals