9 use rsem_perl_utils qw(runCommand collectResults showVersionInfo);
15 my $CONFIDENCE = 0.95;
18 my $NMB = 1024; # default
22 my $read_type = 1; # default, single end with qual
29 my $chunkMbs = 0; # 0 = use bowtie default
50 my $genBamF = 1; # default is generating transcript bam file
51 my $genGenomeBamF = 0;
54 my $var_opt = 0; # temporarily, only for internal use
60 my $keep_intermediate_files = 0;
62 my $strand_specific = 0;
65 my $bowtie2_path = "";
66 my $bowtie2_mismatch_rate = 0.1;
68 my $bowtie2_sensitivity_level = "sensitive"; # must be one of "very_fast", "fast", "sensitive", "very_sensitive"
73 my ($time_start, $time_end, $time_alignment, $time_rsem, $time_ci) = (0, 0, 0, 0, 0);
79 my ($refName, $sampleName, $sampleToken, $temp_dir, $stat_dir, $imdName, $statName) = ();
82 GetOptions("keep-intermediate-files" => \$keep_intermediate_files,
83 "temporary-folder=s" => \$temp_dir,
84 "no-qualities" => \$no_qual,
85 "paired-end" => \$paired_end,
86 "strand-specific" => \$strand_specific,
89 "sam-header-info=s" => \$fn_list,
91 "seed-length=i" => \$L,
92 "bowtie-path=s" => \$bowtie_path,
95 "bowtie-m=i" => \$maxHits,
96 "bowtie-chunkmbs=i" => \$chunkMbs,
97 "phred33-quals" => \$phred33,
98 "phred64-quals" => \$phred64, #solexa1.3-quals" => \$phred64,
99 "solexa-quals" => \$solexa,
100 "bowtie2" => \$bowtie2,
101 "bowtie2-path=s" => \$bowtie2_path,
102 "bowtie2-mismatch-rate=f" => \$bowtie2_mismatch_rate,
103 "bowtie2-k=i" => \$bowtie2_k,
104 "bowtie2-sensitivity-level=s" => \$bowtie2_sensitivity_level,
105 "forward-prob=f" => \$probF,
106 "fragment-length-min=i" => \$minL,
107 "fragment-length-max=i" => \$maxL,
108 "fragment-length-mean=f" => \$mean,
109 "fragment-length-sd=f" => \$sd,
110 "estimate-rspd" => \$estRSPD,
111 "num-rspd-bins=i" => \$B,
112 "p|num-threads=i" => \$nThreads,
113 "no-bam-output" => sub { $genBamF = 0; },
114 "output-genome-bam" => \$genGenomeBamF,
115 "sampling-for-bam" => \$sampling,
117 "calc-ci" => \$calcCI,
118 "ci-memory=i" => \$NMB,
120 "version" => \$version,
121 "q|quiet" => \$quiet,
122 "h|help" => \$help) or pod2usage(-exitval => 2, -verbose => 2);
124 my $dir = "$FindBin::Bin/";
126 pod2usage(-verbose => 2) if ($help == 1);
127 &showVersionInfo($dir) if ($version == 1);
129 #check parameters and options
131 if ($is_sam || $is_bam) {
132 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (scalar(@ARGV) != 3);
133 pod2usage(-msg => "--sam and --bam cannot be active at the same time!", -exitval => 2, -verbose => 2) if ($is_sam == 1&& $is_bam == 1);
134 pod2usage(-msg => "--bowtie-path, --bowtie-n, --bowtie-e, --bowtie-m, --phred33-quals, --phred64-quals, --solexa-quals, --bowtie2, --bowtie2-path, --bowtie2-mismatch-rate, --bowtie2-k and --bowtie2-sensitivity-level cannot be set if input is SAM/BAM format!", -exitval => 2, -verbose => 2) if ($bowtie_path ne "" || $C != 2 || $E != 99999999 || $maxHits != 200 || $phred33 || $phred64 || $solexa || $bowtie2 || $bowtie2_path ne "" || $bowtie2_mismatch_rate != 0.1 || $bowtie2_k != 200 || $bowtie2_sensitivity_level ne "sensitive");
137 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (!$paired_end && scalar(@ARGV) != 3 || $paired_end && scalar(@ARGV) != 4);
138 pod2usage(-msg => "If --no-qualities is set, neither --phred33-quals, --phred64-quals or --solexa-quals can be active!", -exitval => 2, -verbose => 2) if ($no_qual && ($phred33 + $phred64 + $solexa > 0));
139 pod2usage(-msg => "Only one of --phred33-quals, --phred64-quals, and --solexa-quals can be active!", -exitval => 2, -verbose => 2) if ($phred33 + $phred64 + $solexa > 1);
140 pod2usage(-msg => "--sam , --bam or --sam-header-info cannot be set if use bowtie/bowtie2 aligner to produce alignments!", -exitval => 2, -verbose => 2) if ($is_sam || $is_bam || $fn_list ne "");
141 pod2usage(-msg => "--bowtie2-path, --bowtie2-mismatch-rate, --bowtie2-k and --bowtie2-sensitivity-level cannot be set if bowtie aligner is used!", -exitval => 2, -verbose => 2) if (!$bowtie2 && ($bowtie2_path ne "" || $bowtie2_mismatch_rate != 0.1 || $bowtie2_k != 200 || $bowtie2_sensitivity_level ne "sensitive"));
142 pod2usage(-msg => "--bowtie-path, --bowtie-n, --bowtie-e, --bowtie-m cannot be set if bowtie2 aligner is used!", -exitval => 2, -verbose => 2) if ($bowtie2 && ($bowtie_path ne "" || $C != 2 || $E != 99999999 || $maxHits != 200));
143 pod2usage(-msg => "Mismatch rate must be within [0, 1]!", -exitval => 2, -verbose => 2) if ($bowtie2 && ($bowtie2_mismatch_rate < 0.0 || $bowtie2_mismatch_rate > 1.0));
144 pod2usage(-msg => "Sensitivity level must be one of \"very_fast\", \"fast\", \"sensitive\", and \"very_sensitive\"!", -exitval => 2, -verbose => 2) if ($bowtie2 && (($bowtie2_sensitivity_level ne "very_fast") && ($bowtie2_sensitivity_level ne "fast") && ($bowtie2_sensitivity_level ne "sensitive") && ($bowtie2_sensitivity_level ne "very_sensitive")));
147 pod2usage(-msg => "Forward probability should be in [0, 1]!", -exitval => 2, -verbose => 2) if ($probF < 0 || $probF > 1);
148 pod2usage(-msg => "Min fragment length should be at least 1!", -exitval => 2, -verbose => 2) if ($minL < 1);
149 pod2usage(-msg => "Min fragment length should be smaller or equal to max fragment length!", -exitval => 2, -verbose => 2) if ($minL > $maxL);
150 pod2usage(-msg => "The memory allocated for calculating credibility intervals should be at least 1 MB!\n", -exitval => 2, -verbose => 2) if ($NMB < 1);
151 pod2usage(-msg => "Number of threads should be at least 1!\n", -exitval => 2, -verbose => 2) if ($nThreads < 1);
152 pod2usage(-msg => "Seed length should be at least 5!\n", -exitval => 2, -verbose => 2) if ($L < 5);
153 pod2usage(-msg => "--sampling-for-bam cannot be specified if --no-bam-output is specified!\n", -exitval => 2, -verbose => 2) if ($sampling && !$genBamF);
154 pod2usage(-msg => "--output-genome-bam cannot be specified if --no-bam-output is specified!\n", -exitval => 2, -verbose => 2) if ($genGenomeBamF && !$genBamF);
156 if ($L < 25) { print "Warning: the seed length set is less than 25! This is only allowed if the references are not added poly(A) tails.\n"; }
158 if ($strand_specific) { $probF = 1.0; }
161 if ($no_qual) { $read_type = 2; }
162 else { $read_type = 3; }
165 if ($no_qual) { $read_type = 0; }
166 else { $read_type = 1; }
169 if (scalar(@ARGV) == 3) {
170 if ($is_sam || $is_bam) { $inpF = $ARGV[0]; }
171 else {$mate1_list = $ARGV[0]; }
173 $sampleName = $ARGV[2];
176 $mate1_list = $ARGV[0];
177 $mate2_list = $ARGV[1];
179 $sampleName = $ARGV[3];
182 if ($genGenomeBamF) {
183 open(INPUT, "$refName.ti");
184 my $line = <INPUT>; chomp($line);
186 my ($M, $type) = split(/ /, $line);
187 pod2usage(-msg => "No genome information provided, so genome bam file cannot be generated!\n", -exitval => 2, -verbose => 2) if ($type != 0);
190 my $pos = rindex($sampleName, '/');
191 if ($pos < 0) { $sampleToken = $sampleName; }
192 else { $sampleToken = substr($sampleName, $pos + 1); }
194 if ($temp_dir eq "") { $temp_dir = "$sampleName.temp"; }
195 $stat_dir = "$sampleName.stat";
197 if (!(-d $temp_dir) && !mkdir($temp_dir)) { print "Fail to create folder $temp_dir.\n"; exit(-1); }
198 if (!(-d $stat_dir) && !mkdir($stat_dir)) { print "Fail to create folder $stat_dir.\n"; exit(-1); }
200 $imdName = "$temp_dir/$sampleToken";
201 $statName = "$stat_dir/$sampleToken";
203 if (!$is_sam && !$is_bam && !$no_qual && ($phred33 + $phred64 + $solexa == 0)) { $phred33 = 1; }
205 my ($mate_minL, $mate_maxL) = (1, $maxL);
207 if ($bowtie_path ne "") { $bowtie_path .= "/"; }
208 if ($bowtie2_path ne "") { $bowtie2_path .= "/"; }
212 if (!$is_sam && !$is_bam) {
214 $command = $bowtie_path."bowtie";
215 if ($no_qual) { $command .= " -f"; }
216 else { $command .= " -q"; }
218 if ($phred33) { $command .= " --phred33-quals"; }
219 elsif ($phred64) { $command .= " --phred64-quals"; }
220 elsif ($solexa) { $command .= " --solexa-quals"; }
222 $command .= " -n $C -e $E -l $L";
223 if ($read_type == 2 || $read_type == 3) { $command .= " -I $minL -X $maxL"; }
224 if ($chunkMbs > 0) { $command .= " --chunkmbs $chunkMbs"; }
226 if ($strand_specific || $probF == 1.0) { $command .= " --norc"; }
227 elsif ($probF == 0.0) { $command .= " --nofw"; }
229 $command .= " -p $nThreads -a -m $maxHits -S";
230 if ($quiet) { $command .= " --quiet"; }
232 $command .= " $refName";
233 if ($read_type == 0 || $read_type == 1) {
234 $command .= " $mate1_list";
237 $command .= " -1 $mate1_list -2 $mate2_list";
240 # pipe to samtools to generate a BAM file
241 $command .= " | $dir\sam/samtools view -S -b -o $imdName.bam -";
244 $command = $bowtie2_path."bowtie2";
245 if ($no_qual) { $command .= " -f"; }
246 else { $command .= " -q"; }
248 if ($phred33) { $command .= " --phred33"; }
249 elsif ($phred64) { $command .= " --phred64"; }
250 elsif ($solexa) { $command .= " --solexa-quals"; }
252 if ($bowtie2_sensitivity_level eq "very_fast") { $command .= " --very-fast"; }
253 elsif ($bowtie2_sensitivity_level eq "fast") { $command .= " --fast"; }
254 elsif ($bowtie2_sensitivity_level eq "sensitive") { $command .= " --sensitive"; }
255 else { $command .= " --very-sensitive"; }
257 $command .= " --dpad 0 --gbar 99999999 --mp 1,1 --np 1 --score-min L,0,-$bowtie2_mismatch_rate";
259 if ($read_type == 2 || $read_type == 3) { $command .= " -I $minL -X $maxL --no-mixed --no-discordant"; }
261 if ($strand_specific || $probF == 1.0) { $command .= " --norc"; }
262 elsif ($probF == 0.0) { $command .= " --nofw"; }
264 $command .= " -p $nThreads -k $bowtie2_k";
265 if ($quiet) { $command .= " --quiet"; }
267 $command .= " -x $refName";
268 if ($read_type == 0 || $read_type == 1) {
269 $command .= " -U $mate1_list";
272 $command .= " -1 $mate1_list -2 $mate2_list";
275 # pipe to samtools to generate a BAM file
276 $command .= " | $dir\sam/samtools view -S -b -o $imdName.bam -";
279 if ($mTime) { $time_start = time(); }
281 &runCommand($command);
283 if ($mTime) { $time_end = time(); $time_alignment = $time_end - $time_start; }
285 $inpF = "$imdName.bam";
286 $is_bam = 1; # alignments are outputed as a BAM file
289 if ($mTime) { $time_start = time(); }
291 $command = $dir."rsem-parse-alignments $refName $imdName $statName";
294 if ($is_sam) { $samInpType = "s"; }
295 elsif ($is_bam) { $samInpType = "b"; }
297 $command .= " $samInpType $inpF -t $read_type";
298 if ($fn_list ne "") { $command .= " -l $fn_list"; }
299 if ($tagName ne "") { $command .= " -tag $tagName"; }
300 if ($quiet) { $command .= " -q"; }
302 &runCommand($command);
304 $command = $dir."rsem-build-read-index $gap";
305 if ($read_type == 0) { $command .= " 0 $quiet $imdName\_alignable.fa"; }
306 elsif ($read_type == 1) { $command .= " 1 $quiet $imdName\_alignable.fq"; }
307 elsif ($read_type == 2) { $command .= " 0 $quiet $imdName\_alignable_1.fa $imdName\_alignable_2.fa"; }
308 elsif ($read_type == 3) { $command .= " 1 $quiet $imdName\_alignable_1.fq $imdName\_alignable_2.fq"; }
309 else { print "Impossible! read_type is not in [1,2,3,4]!\n"; exit(-1); }
310 &runCommand($command);
312 my $doesOpen = open(OUTPUT, ">$imdName.mparams");
313 if ($doesOpen == 0) { print "Cannot generate $imdName.mparams!\n"; exit(-1); }
314 print OUTPUT "$minL $maxL\n";
315 print OUTPUT "$probF\n";
316 print OUTPUT "$estRSPD\n";
318 print OUTPUT "$mate_minL $mate_maxL\n";
319 print OUTPUT "$mean $sd\n";
323 $command = $dir."rsem-run-em $refName $read_type $sampleName $imdName $statName -p $nThreads";
325 $command .= " -b $samInpType $inpF";
326 if ($fn_list ne "") { $command .= " 1 $fn_list"; }
327 else { $command .= " 0"; }
328 if ($sampling) { $command .= " --sampling"; }
330 if ($calcCI || $var_opt) { $command .= " --gibbs-out"; }
331 if ($quiet) { $command .= " -q"; }
333 &runCommand($command);
335 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
336 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
339 $command = $dir."sam/samtools sort $sampleName.transcript.bam $sampleName.transcript.sorted";
340 &runCommand($command);
341 $command = $dir."sam/samtools index $sampleName.transcript.sorted.bam";
342 &runCommand($command);
344 if ($genGenomeBamF) {
345 $command = $dir."rsem-tbam2gbam $refName $sampleName.transcript.bam $sampleName.genome.bam";
346 &runCommand($command);
347 $command = $dir."sam/samtools sort $sampleName.genome.bam $sampleName.genome.sorted";
348 &runCommand($command);
349 $command = $dir."sam/samtools index $sampleName.genome.sorted.bam";
350 &runCommand($command);
354 if ($mTime) { $time_end = time(); $time_rsem = $time_end - $time_start; }
356 if ($mTime) { $time_start = time(); }
358 if ($calcCI || $var_opt) {
359 $command = $dir."rsem-run-gibbs $refName $imdName $statName $BURNIN $NCV $SAMPLEGAP";
360 $command .= " -p $nThreads";
361 if ($var_opt) { $command .= " --var"; }
362 if ($quiet) { $command .= " -q"; }
363 &runCommand($command);
367 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak1");
368 system("mv $sampleName.genes.results $imdName.genes.results.bak1");
369 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
370 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
372 $command = $dir."rsem-calculate-credibility-intervals $refName $imdName $statName $CONFIDENCE $NCV $NSPC $NMB";
373 $command .= " -p $nThreads";
374 if ($quiet) { $command .= " -q"; }
375 &runCommand($command);
377 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak2");
378 system("mv $sampleName.genes.results $imdName.genes.results.bak2");
379 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
380 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
383 if ($mTime) { $time_end = time(); $time_ci = $time_end - $time_start; }
385 if ($mTime) { $time_start = time(); }
387 if (!$keep_intermediate_files) {
388 &runCommand("rm -rf $temp_dir", "Fail to delete the temporary folder!");
391 if ($mTime) { $time_end = time(); }
394 open(OUTPUT, ">$sampleName.time");
395 print OUTPUT "Aligning reads: $time_alignment s.\n";
396 print OUTPUT "Estimating expression levels: $time_rsem s.\n";
397 print OUTPUT "Calculating credibility intervals: $time_ci s.\n";
398 my $time_del = $time_end - $time_start;
399 # print OUTPUT "Delete: $time_del s.\n";
407 rsem-calculate-expression
411 rsem-calculate-expression [options] upstream_read_file(s) reference_name sample_name
412 rsem-calculate-expression [options] --paired-end upstream_read_file(s) downstream_read_file(s) reference_name sample_name
413 rsem-calculate-expression [options] --sam/--bam [--paired-end] input reference_name sample_name
419 =item B<upstream_read_files(s)>
421 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.
423 =item B<downstream_read_file(s)>
425 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.
429 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.
431 =item B<reference_name>
433 The name of the reference used. The user must have run 'rsem-prepare-reference' with this reference_name before running this program.
437 The name of the sample analyzed. All output files are prefixed by this name (e.g., sample_name.genes.results)
445 =item B<--paired-end>
447 Input reads are paired-end reads. (Default: off)
449 =item B<--no-qualities>
451 Input reads do not contain quality scores. (Default: off)
453 =item B<--strand-specific>
455 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)
459 Input file is in SAM format. (Default: off)
463 Input file is in BAM format. (Default: off)
465 =item B<--sam-header-info> <file>
467 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: "")
469 =item B<-p/--num-threads> <int>
471 Number of threads to use. Both Bowtie/Bowtie2 and expression estimation will use this many threads. (Default: 1)
473 =item B<--no-bam-output>
475 Do not output any BAM file. (Default: off)
477 =item B<--output-genome-bam>
479 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)
481 =item B<--sampling-for-bam>
483 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)
487 Calculate 95% credibility intervals and posterior mean estimates. (Default: off)
489 =item B<--seed-length> <int>
491 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)
493 =item B<--tag> <string>
495 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: "")
497 =item B<--bowtie-path> <path>
499 The path to the Bowtie executables. (Default: the path to the Bowtie executables is assumed to be in the user's PATH environment variable)
501 =item B<--bowtie-n> <int>
503 (Bowtie parameter) max # of mismatches in the seed. (Range: 0-3, Default: 2)
505 =item B<--bowtie-e> <int>
507 (Bowtie parameter) max sum of mismatch quality scores across the alignment. (Default: 99999999)
509 =item B<--bowtie-m> <int>
511 (Bowtie parameter) suppress all alignments for a read if > <int> valid alignments exist. (Default: 200)
513 =item B<--bowtie-chunkmbs> <int>
515 (Bowtie parameter) memory allocated for best first alignment calculation (Default: 0 - use Bowtie's default)
517 =item B<--phred33-quals>
519 Input quality scores are encoded as Phred+33. (Default: on)
521 =item B<--phred64-quals>
523 Input quality scores are encoded as Phred+64 (default for GA Pipeline ver. >= 1.3). (Default: off)
525 =item B<--solexa-quals>
527 Input quality scores are solexa encoded (from GA Pipeline ver. < 1.3). (Default: off)
531 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 '--very-sensitive --dpad 0 --gbar 99999999 --mp 1,1 --np 1 --score-min L,0,-rate'. "-rate", the last parameter of '--score-min' is the negative value of the mismatch rate provided by option '--bowtie2-mismatch-rate'. If reads are paired-end, we additionally use options '--no-mixed' and '--no-discordant'. (Default: off)
533 =item B<--bowtie2-path> <path>
535 (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)
537 =item B<--bowtie2-mismatch-rate> <double>
539 (Bowtie 2 parameter) The maximum mismatch rate allowed. (Default: 0.1)
541 =item B<--bowtie2-k> <int>
543 (Bowtie 2 parameter) Find up to <int> alignments per read. (Default: 200)
545 =item B<--bowtie2-sensitivity-level> <string>
547 (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)
549 =item B<--forward-prob> <double>
551 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)
553 =item B<--fragment-length-min> <int>
555 Minimum read/insert length allowed. This is also the value for the Bowtie/Bowtie2 -I option. (Default: 1)
557 =item B<--fragment-length-max> <int>
559 Maximum read/insert length allowed. This is also the value for the Bowtie/Bowtie 2 -X option. (Default: 1000)
561 =item B<--fragment-length-mean> <double>
563 (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)
565 =item B<--fragment-length-sd> <double>
567 (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>)
569 =item B<--estimate-rspd>
571 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)
573 =item B<--num-rspd-bins> <int>
575 Number of bins in the RSPD. Only relevant when '--estimate-rspd' is specified. Use of the default setting is recommended. (Default: 20)
577 =item B<--ci-memory> <int>
579 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)
581 =item B<--keep-intermediate-files>
583 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)
585 =item B<--temporary-folder> <string>
587 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)
591 Output time consumed by each step of RSEM to 'sample_name.time'. (Default: off)
595 Suppress the output of logging information. (Default: off)
599 Show help information.
603 Show version information.
609 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.
611 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:
613 convert-sam-for-rsem /ref/mouse_125 input.sam -o input_for_rsem.sam
615 For details, please refer to 'convert-sam-for-rsem's documentation page.
617 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 '*'.
619 The user must run 'rsem-prepare-reference' with the appropriate reference before using this program.
621 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.
623 Please note that some of the default values for the Bowtie parameters are not the same as those defined for Bowtie itself.
625 The temporary directory and all intermediate files will be removed when RSEM finishes unless '--keep-intermediate-files' is specified.
627 With the '--calc-ci' option, 95% credibility intervals and posterior mean estimates will be calculated in addition to maximum likelihood estimates.
633 =item B<sample_name.isoforms.results>
635 File containing isoform level expression estimates. The first line
636 contains column names separated by the tab character. The format of
637 each line in the rest of this file is:
639 transcript_id gene_id length effective_length expected_count TPM FPKM IsoPct [pme_expected_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]
641 Fields are separated by the tab character. Fields within "[]" are only
642 presented if '--calc-ci' is set.
644 'transcript_id' is the transcript name of this transcript. 'gene_id'
645 is the gene name of the gene which this transcript belongs to (denote
646 this gene as its parent gene). If no gene information is provided,
647 'gene_id' and 'transcript_id' are the same.
649 'length' is this transcript's sequence length (poly(A) tail is not
650 counted). 'effective_length' counts only the positions that can
651 generate a valid fragment. If no poly(A) tail is added,
652 'effective_length' is equal to transcript length - mean fragment
653 length + 1. If one transcript's effective length is less than 1, this
654 transcript's both effective length and abundance estimates are set to
657 'expected_count' is the sum of the posterior probability of each read
658 comes from this transcript over all reads. Because 1) each read
659 aligning to this transcript has a probability of being generated from
660 background noise; 2) RSEM may filter some alignable low quality reads,
661 the sum of expected counts for all transcript are generally less than
662 the total number of reads aligned.
664 'TPM' stands for Transcripts Per Million. It is a relative measure of
665 transcript abundance. The sum of all transcripts' TPM is 1
666 million. 'FPKM' stands for Fragments Per Kilobase of transcript per
667 Million mapped reads. It is another relative measure of transcript
668 abundance. If we define l_bar be the mean transcript length in a
669 sample, which can be calculated as
671 l_bar = \sum_i TPM_i / 10^6 * effective_length_i (i goes through every transcript),
673 the following equation is hold:
675 FPKM_i = 10^3 / l_bar * TPM_i.
677 We can see that the sum of FPKM is not a constant across samples.
679 'IsoPct' stands for isoform percentage. It is the percentage of this
680 transcript's abandunce over its parent gene's abandunce. If its parent
681 gene has only one isoform or the gene information is not provided,
682 this field will be set to 100.
684 'pme_expected_count', 'pme_TPM', 'pme_FPKM' are posterior mean
685 estimates calculated by RSEM's Gibbs sampler. 'IsoPct_from_pme_TPM' is
686 the isoform percentage calculated from 'pme_TPM' values.
688 'TPM_ci_lower_bound', 'TPM_ci_upper_bound', 'FPKM_ci_lower_bound' and
689 'FPKM_ci_upper_bound' are lower(l) and upper(u) bounds of 95%
690 credibility intervals for TPM and FPKM values. The bounds are
691 inclusive (i.e. [l, u]).
693 =item B<sample_name.genes.results>
695 File containing gene level expression estimates. The first line
696 contains column names separated by the tab character. The format of
697 each line in the rest of this file is:
699 gene_id transcript_id(s) length effective_length expected_count TPM FPKM [pme_expected_count pme_TPM pme_FPKM TPM_ci_lower_bound TPM_ci_upper_bound FPKM_ci_lower_bound FPKM_ci_upper_bound]
701 Fields are separated by the tab character. Fields within "[]" are only
702 presented if '--calc-ci' is set.
704 'transcript_id(s)' is a comma-separated list of transcript_ids
705 belonging to this gene. If no gene information is provided, 'gene_id'
706 and 'transcript_id(s)' are identical (the 'transcript_id').
708 A gene's 'length' and 'effective_length' are
709 defined as the weighted average of its transcripts' lengths and
710 effective lengths (weighted by 'IsoPct'). A gene's abundance estimates
711 are just the sum of its transcripts' abundance estimates.
713 =item B<sample_name.transcript.bam, sample_name.transcript.sorted.bam and sample_name.transcript.sorted.bam.bai>
715 Only generated when --no-bam-output is not specified.
717 'sample_name.transcript.bam' is a BAM-formatted file of read
718 alignments in transcript coordinates. The MAPQ field of each alignment
719 is set to min(100, floor(-10 * log10(1.0 - w) + 0.5)), where w is the
720 posterior probability of that alignment being the true mapping of a
721 read. In addition, RSEM pads a new tag ZW:f:value, where value is a
722 single precision floating number representing the posterior
723 probability. Because this file contains all alignment lines produced
724 by bowtie or user-specified aligners, it can also be used as a
725 replacement of the aligner generated BAM/SAM file. For paired-end
726 reads, if one mate has alignments but the other does not, this file
727 marks the alignable mate as "unmappable" (flag bit 0x4) and appends an
728 optional field "Z0:A:!".
730 'sample_name.transcript.sorted.bam' and
731 'sample_name.transcript.sorted.bam.bai' are the sorted BAM file and
732 indices generated by samtools (included in RSEM package).
734 =item B<sample_name.genome.bam, sample_name.genome.sorted.bam and sample_name.genome.sorted.bam.bai>
736 Only generated when --no-bam-output is not specified and --output-genome-bam is specified.
738 'sample_name.genome.bam' is a BAM-formatted file of read alignments in
739 genomic coordinates. Alignments of reads that have identical genomic
740 coordinates (i.e., alignments to different isoforms that share the
741 same genomic region) are collapsed into one alignment. The MAPQ field
742 of each alignment is set to min(100, floor(-10 * log10(1.0 - w) +
743 0.5)), where w is the posterior probability of that alignment being
744 the true mapping of a read. In addition, RSEM pads a new tag
745 ZW:f:value, where value is a single precision floating number
746 representing the posterior probability. If an alignment is spliced, a
747 XS:A:value tag is also added, where value is either '+' or '-'
748 indicating the strand of the transcript it aligns to.
750 'sample_name.genome.sorted.bam' and 'sample_name.genome.sorted.bam.bai' are the
751 sorted BAM file and indices generated by samtools (included in RSEM package).
753 =item B<sample_name.time>
755 Only generated when --time is specified.
757 It contains time (in seconds) consumed by aligning reads, estimating expression levels and calculating credibility intervals.
759 =item B<sample_name.stat>
761 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.
767 Assume the path to the bowtie executables is in the user's PATH environment variable. Reference files are under '/ref' with name 'mouse_125'.
769 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:
771 rsem-calculate-expression --phred64-quals \
773 --output-genome-bam \
778 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:
780 rsem-calculate-expression -p 8 \
785 mmliver_paired_end_quals
787 3) '/data/mmliver.fa', single-end reads without quality scores. We want to use 8 threads:
789 rsem-calculate-expression -p 8 \
793 mmliver_single_without_quals
795 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:
797 rsem-calculate-expression --bowtie-path /sw/bowtie \
799 --fragment-length-mean 150.0 \
800 --fragment-length-sd 35.0 \
802 --output-genome-bam \
809 5) '/data/mmliver_paired_end_quals.bam', paired-end reads with quality scores. We want to use 8 threads:
811 rsem-calculate-expression --paired-end \
814 /data/mmliver_paired_end_quals.bam \
816 mmliver_paired_end_quals