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 $strand_specific = 0;
67 my ($time_start, $time_end, $time_alignment, $time_rsem, $time_ci) = (0, 0, 0, 0, 0);
73 my ($refName, $sampleName, $sampleToken, $temp_dir, $stat_dir, $imdName, $statName) = ();
76 GetOptions("keep-intermediate-files" => \$keep_intermediate_files,
77 "temporary-folder=s" => \$temp_dir,
78 "no-qualities" => \$no_qual,
79 "paired-end" => \$paired_end,
80 "strand-specific" => \$strand_specific,
83 "sam-header-info=s" => \$fn_list,
85 "seed-length=i" => \$L,
86 "bowtie-path=s" => \$bowtie_path,
89 "bowtie-m=i" => \$maxHits,
90 "bowtie-chunkmbs=i" => \$chunkMbs,
91 "bowtie-phred33-quals" => \$phred33,
92 "bowtie-phred64-quals" => \$phred64, #solexa1.3-quals" => \$phred64,
93 "bowtie-solexa-quals" => \$solexa,
94 "forward-prob=f" => \$probF,
95 "fragment-length-min=i" => \$minL,
96 "fragment-length-max=i" => \$maxL,
97 "fragment-length-mean=f" => \$mean,
98 "fragment-length-sd=f" => \$sd,
99 "estimate-rspd" => \$estRSPD,
100 "num-rspd-bins=i" => \$B,
101 "p|num-threads=i" => \$nThreads,
102 "no-bam-output" => sub { $genBamF = 0; },
103 "output-genome-bam" => \$genGenomeBamF,
104 "sampling-for-bam" => \$sampling,
106 "calc-ci" => \$calcCI,
107 "ci-memory=i" => \$NMB,
109 "version" => \$version,
110 "q|quiet" => \$quiet,
111 "h|help" => \$help) or pod2usage(-exitval => 2, -verbose => 2);
113 my $dir = "$FindBin::Bin/";
115 pod2usage(-verbose => 2) if ($help == 1);
116 &showVersionInfo($dir) if ($version == 1);
118 #check parameters and options
120 if ($is_sam || $is_bam) {
121 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (scalar(@ARGV) != 3);
122 pod2usage(-msg => "--sam and --bam cannot be active at the same time!", -exitval => 2, -verbose => 2) if ($is_sam == 1&& $is_bam == 1);
123 pod2usage(-msg => "--bowtie-path, --bowtie-n, --bowtie-e, --bowtie-m, --phred33-quals, --phred64-quals or --solexa-quals cannot be set if input is SAM/BAM format!", -exitval => 2, -verbose => 2) if ($bowtie_path ne "" || $C != 2 || $E != 99999999 || $maxHits != 200 || $phred33 || $phred64 || $solexa);
126 pod2usage(-msg => "Invalid number of arguments!", -exitval => 2, -verbose => 2) if (!$paired_end && scalar(@ARGV) != 3 || $paired_end && scalar(@ARGV) != 4);
127 pod2usage(-msg => "Only one of --phred33-quals --phred64-quals/--solexa1.3-quals --solexa-suqls can be active!", -exitval => 2, -verbose => 2) if ($phred33 + $phred64 + $solexa > 1);
128 podwusage(-msg => "--sam , --bam or --sam-header-info cannot be set if use bowtie aligner to produce alignments!", -exitval => 2, -verbose => 2) if ($is_sam || $is_bam || $fn_list ne "");
131 pod2usage(-msg => "Forward probability should be in [0, 1]!", -exitval => 2, -verbose => 2) if ($probF < 0 || $probF > 1);
132 pod2usage(-msg => "Min fragment length should be at least 1!", -exitval => 2, -verbose => 2) if ($minL < 1);
133 pod2usage(-msg => "Min fragment length should be smaller or equal to max fragment length!", -exitval => 2, -verbose => 2) if ($minL > $maxL);
134 pod2usage(-msg => "The memory allocated for calculating credibility intervals should be at least 1 MB!\n", -exitval => 2, -verbose => 2) if ($NMB < 1);
135 pod2usage(-msg => "Number of threads should be at least 1!\n", -exitval => 2, -verbose => 2) if ($nThreads < 1);
136 pod2usage(-msg => "Seed length should be at least 5!\n", -exitval => 2, -verbose => 2) if ($L < 5);
137 pod2usage(-msg => "--sampling-for-bam cannot be specified if --no-bam-output is specified!\n", -exitval => 2, -verbose => 2) if ($sampling && !$genBamF);
138 pod2usage(-msg => "--output-genome-bam cannot be specified if --no-bam-output is specified!\n", -exitval => 2, -verbose => 2) if ($genGenomeBamF && !$genBamF);
140 if ($L < 25) { print "Warning: the seed length set is less than 25! This is only allowed if the references are not added poly(A) tails.\n"; }
142 if ($strand_specific) { $probF = 1.0; }
145 if ($no_qual) { $read_type = 2; }
146 else { $read_type = 3; }
149 if ($no_qual) { $read_type = 0; }
150 else { $read_type = 1; }
153 if (scalar(@ARGV) == 3) {
154 if ($is_sam || $is_bam) { $inpF = $ARGV[0]; }
155 else {$mate1_list = $ARGV[0]; }
157 $sampleName = $ARGV[2];
160 $mate1_list = $ARGV[0];
161 $mate2_list = $ARGV[1];
163 $sampleName = $ARGV[3];
166 if ($genGenomeBamF) {
167 open(INPUT, "$refName.ti");
168 my $line = <INPUT>; chomp($line);
170 my ($M, $type) = split(/ /, $line);
171 pod2usage(-msg => "No genome information provided, so genome bam file cannot be generated!\n", -exitval => 2, -verbose => 2) if ($type != 0);
174 my $pos = rindex($sampleName, '/');
175 if ($pos < 0) { $sampleToken = $sampleName; }
176 else { $sampleToken = substr($sampleName, $pos + 1); }
178 if ($temp_dir eq "") { $temp_dir = "$sampleName.temp"; }
179 $stat_dir = "$sampleName.stat";
181 if (!(-d $temp_dir) && !mkdir($temp_dir)) { print "Fail to create folder $temp_dir.\n"; exit(-1); }
182 if (!(-d $stat_dir) && !mkdir($stat_dir)) { print "Fail to create folder $stat_dir.\n"; exit(-1); }
184 $imdName = "$temp_dir/$sampleToken";
185 $statName = "$stat_dir/$sampleToken";
187 if (!$is_sam && !$is_bam && !$no_qual && ($phred33 + $phred64 + $solexa == 0)) { $phred33 = 1; }
189 my ($mate_minL, $mate_maxL) = (1, $maxL);
191 if ($bowtie_path ne "") { $bowtie_path .= "/"; }
195 if (!$is_sam && !$is_bam) {
196 $command = $bowtie_path."bowtie";
197 if ($no_qual) { $command .= " -f"; }
198 else { $command .= " -q"; }
200 if ($phred33) { $command .= " --phred33-quals"; }
201 elsif ($phred64) { $command .= " --phred64-quals"; }
202 elsif ($solexa) { $command .= " --solexa-quals"; }
204 $command .= " -n $C -e $E -l $L";
205 if ($read_type == 2 || $read_type == 3) { $command .= " -I $minL -X $maxL"; }
206 if ($chunkMbs > 0) { $command .= " --chunkmbs $chunkMbs"; }
208 if ($strand_specific || $probF == 1.0) { $command .= " --norc"; }
209 elsif ($probF == 0.0) { $command .= " --nofw"; }
211 $command .= " -p $nThreads -a -m $maxHits -S";
212 if ($quiet) { $command .= " --quiet"; }
214 $command .= " $refName";
215 if ($read_type == 0 || $read_type == 1) {
216 $command .= " $mate1_list";
219 $command .= " -1 $mate1_list -2 $mate2_list";
222 # pipe to samtools to generate a BAM file
223 $command .= " | $dir\sam/samtools view -S -b -o $imdName.bam -";
225 if ($mTime) { $time_start = time(); }
227 &runCommand($command);
229 if ($mTime) { $time_end = time(); $time_alignment = $time_end - $time_start; }
231 $inpF = "$imdName.bam";
232 $is_bam = 1; # alignments are outputed as a BAM file
235 if ($mTime) { $time_start = time(); }
237 $command = $dir."rsem-parse-alignments $refName $imdName $statName";
240 if ($is_sam) { $samInpType = "s"; }
241 elsif ($is_bam) { $samInpType = "b"; }
243 $command .= " $samInpType $inpF -t $read_type";
244 if ($fn_list ne "") { $command .= " -l $fn_list"; }
245 if ($tagName ne "") { $command .= " -tag $tagName"; }
246 if ($quiet) { $command .= " -q"; }
248 &runCommand($command);
250 $command = $dir."rsem-build-read-index $gap";
251 if ($read_type == 0) { $command .= " 0 $quiet $imdName\_alignable.fa"; }
252 elsif ($read_type == 1) { $command .= " 1 $quiet $imdName\_alignable.fq"; }
253 elsif ($read_type == 2) { $command .= " 0 $quiet $imdName\_alignable_1.fa $imdName\_alignable_2.fa"; }
254 elsif ($read_type == 3) { $command .= " 1 $quiet $imdName\_alignable_1.fq $imdName\_alignable_2.fq"; }
255 else { print "Impossible! read_type is not in [1,2,3,4]!\n"; exit(-1); }
256 &runCommand($command);
258 my $doesOpen = open(OUTPUT, ">$imdName.mparams");
259 if ($doesOpen == 0) { print "Cannot generate $imdName.mparams!\n"; exit(-1); }
260 print OUTPUT "$minL $maxL\n";
261 print OUTPUT "$probF\n";
262 print OUTPUT "$estRSPD\n";
264 print OUTPUT "$mate_minL $mate_maxL\n";
265 print OUTPUT "$mean $sd\n";
269 $command = $dir."rsem-run-em $refName $read_type $sampleName $imdName $statName -p $nThreads";
271 $command .= " -b $samInpType $inpF";
272 if ($fn_list ne "") { $command .= " 1 $fn_list"; }
273 else { $command .= " 0"; }
274 if ($sampling) { $command .= " --sampling"; }
276 if ($calcCI || $var_opt) { $command .= " --gibbs-out"; }
277 if ($quiet) { $command .= " -q"; }
279 &runCommand($command);
281 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
282 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
285 $command = $dir."sam/samtools sort $sampleName.transcript.bam $sampleName.transcript.sorted";
286 &runCommand($command);
287 $command = $dir."sam/samtools index $sampleName.transcript.sorted.bam";
288 &runCommand($command);
290 if ($genGenomeBamF) {
291 $command = $dir."rsem-tbam2gbam $refName $sampleName.transcript.bam $sampleName.genome.bam";
292 &runCommand($command);
293 $command = $dir."sam/samtools sort $sampleName.genome.bam $sampleName.genome.sorted";
294 &runCommand($command);
295 $command = $dir."sam/samtools index $sampleName.genome.sorted.bam";
296 &runCommand($command);
300 if ($mTime) { $time_end = time(); $time_rsem = $time_end - $time_start; }
302 if ($mTime) { $time_start = time(); }
304 if ($calcCI || $var_opt) {
305 $command = $dir."rsem-run-gibbs $refName $imdName $statName $BURNIN $NCV $SAMPLEGAP";
306 $command .= " -p $nThreads";
307 if ($var_opt) { $command .= " --var"; }
308 if ($quiet) { $command .= " -q"; }
309 &runCommand($command);
313 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak1");
314 system("mv $sampleName.genes.results $imdName.genes.results.bak1");
315 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
316 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
318 $command = $dir."rsem-calculate-credibility-intervals $refName $imdName $statName $CONFIDENCE $NCV $NSPC $NMB";
319 $command .= " -p $nThreads";
320 if ($quiet) { $command .= " -q"; }
321 &runCommand($command);
323 system("mv $sampleName.isoforms.results $imdName.isoforms.results.bak2");
324 system("mv $sampleName.genes.results $imdName.genes.results.bak2");
325 &collectResults("isoform", "$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
326 &collectResults("gene", "$imdName.gene_res", "$sampleName.genes.results"); # gene level
329 if ($mTime) { $time_end = time(); $time_ci = $time_end - $time_start; }
331 if ($mTime) { $time_start = time(); }
333 if (!$keep_intermediate_files) {
334 &runCommand("rm -rf $temp_dir", "Fail to delete the temporary folder!");
337 if ($mTime) { $time_end = time(); }
340 open(OUTPUT, ">$sampleName.time");
341 print OUTPUT "Aligning reads: $time_alignment s.\n";
342 print OUTPUT "Estimating expression levels: $time_rsem s.\n";
343 print OUTPUT "Calculating credibility intervals: $time_ci s.\n";
344 my $time_del = $time_end - $time_start;
345 # print OUTPUT "Delete: $time_del s.\n";
353 rsem-calculate-expression
357 rsem-calculate-expression [options] upstream_read_file(s) reference_name sample_name
358 rsem-calculate-expression [options] --paired-end upstream_read_file(s) downstream_read_file(s) reference_name sample_name
359 rsem-calculate-expression [options] --sam/--bam [--paired-end] input reference_name sample_name
365 =item B<upstream_read_files(s)>
367 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.
369 =item B<downstream_read_file(s)>
371 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.
375 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.
377 =item B<reference_name>
379 The name of the reference used. The user must have run 'rsem-prepare-reference' with this reference_name before running this program.
383 The name of the sample analyzed. All output files are prefixed by this name (e.g., sample_name.genes.results)
391 =item B<--paired-end>
393 Input reads are paired-end reads. (Default: off)
395 =item B<--no-qualities>
397 Input reads do not contain quality scores. (Default: off)
399 =item B<--strand-specific>
401 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)
405 Input file is in SAM format. (Default: off)
409 Input file is in BAM format. (Default: off)
411 =item B<--sam-header-info> <file>
413 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: "")
415 =item B<-p/--num-threads> <int>
417 Number of threads to use. Both Bowtie and expression estimation will use this many threads. (Default: 1)
419 =item B<--no-bam-output>
421 Do not output any BAM file. (Default: off)
423 =item B<--output-genome-bam>
425 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)
427 =item B<--sampling-for-bam>
429 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)
433 Calculate 95% credibility intervals and posterior mean estimates. (Default: off)
435 =item B<--seed-length> <int>
437 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)
439 =item B<--tag> <string>
441 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: "")
443 =item B<--bowtie-path> <path>
445 The path to the bowtie executables. (Default: the path to the bowtie executables is assumed to be in the user's PATH environment variable)
447 =item B<--bowtie-n> <int>
449 (Bowtie parameter) max # of mismatches in the seed. (Range: 0-3, Default: 2)
451 =item B<--bowtie-e> <int>
453 (Bowtie parameter) max sum of mismatch quality scores across the alignment. (Default: 99999999)
455 =item B<--bowtie-m> <int>
457 (Bowtie parameter) suppress all alignments for a read if > <int> valid alignments exist. (Default: 200)
459 =item B<--bowtie-chunkmbs> <int>
461 (Bowtie parameter) memory allocated for best first alignment calculation (Default: 0 - use bowtie's default)
463 =item B<--bowtie-phred33-quals>
465 (Bowtie parameter) Input quality scores are encoded as Phred+33. (Default: on)
467 =item B<--bowtie-phred64-quals>
469 (Bowtie parameter) Input quality scores are encoded as Phred+64 (default for GA Pipeline ver. >= 1.3). (Default: off)
471 =item B<--bowtie-solexa-quals>
473 (Bowtie parameter) Input quality scores are solexa encoded (from GA Pipeline ver. < 1.3). (Default: off)
475 =item B<--forward-prob> <double>
477 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)
479 =item B<--fragment-length-min> <int>
481 Minimum read/insert length allowed. This is also the value for the bowtie -I option. (Default: 1)
483 =item B<--fragment-length-max> <int>
485 Maximum read/insert length allowed. This is also the value for the bowtie -X option. (Default: 1000)
487 =item B<--fragment-length-mean> <double>
489 (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)
491 =item B<--fragment-length-sd> <double>
493 (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>)
495 =item B<--estimate-rspd>
497 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)
499 =item B<--num-rspd-bins> <int>
501 Number of bins in the RSPD. Only relevant when '--estimate-rspd' is specified. Use of the default setting is recommended. (Default: 20)
503 =item B<--ci-memory> <int>
505 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)
507 =item B<--keep-intermediate-files>
509 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)
511 =item B<--temporary-folder> <string>
513 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)
517 Output time consumed by each step of RSEM to 'sample_name.time'. (Default: off)
521 Suppress the output of logging information. (Default: off)
525 Show help information.
531 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.
533 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:
535 convert-sam-for-rsem /ref/mouse_125 input.sam -o input_for_rsem.sam
537 For details, please refer to 'convert-sam-for-rsem's documentation page.
539 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 '*'.
541 The user must run 'rsem-prepare-reference' with the appropriate reference before using this program.
543 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.
545 Please note that some of the default values for the Bowtie parameters are not the same as those defined for Bowtie itself.
547 The temporary directory and all intermediate files will be removed when RSEM finishes unless '--keep-intermediate-files' is specified.
549 With the '--calc-ci' option, 95% credibility intervals and posterior mean estimates will be calculated in addition to maximum likelihood estimates.
555 =item B<sample_name.isoforms.results>
557 File containing isoform level expression estimates. The first line
558 contains column names separated by the tab character. The format of
559 each line in the rest of this file is:
561 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]
563 Fields are separated by the tab character. Fields within "[]" are only
564 presented if '--calc-ci' is set.
566 'transcript_id' is the transcript name of this transcript. 'gene_id'
567 is the gene name of the gene which this transcript belongs to (denote
568 this gene as its parent gene). If no gene information is provided,
569 'gene_id' and 'transcript_id' are the same.
571 'length' is this transcript's sequence length (poly(A) tail is not
572 counted). 'effective_length' counts only the positions that can
573 generate a valid fragment. If no poly(A) tail is added,
574 'effective_length' is equal to transcript length - mean fragment
575 length + 1. If one transcript's effective length is less than 1, this
576 transcript's both effective length and abundance estimates are set to
579 'expected_count' is the sum of the posterior probability of each read
580 comes from this transcript over all reads. Because 1) each read
581 aligning to this transcript has a probability of being generated from
582 background noise; 2) RSEM may filter some alignable low quality reads,
583 the sum of expected counts for all transcript are generally less than
584 the total number of reads aligned.
586 'TPM' stands for Transcripts Per Million. It is a relative measure of
587 transcript abundance. The sum of all transcripts' TPM is 1
588 million. 'FPKM' stands for Fragments Per Kilobase of transcript per
589 Million mapped reads. It is another relative measure of transcript
590 abundance. If we define l_bar be the mean transcript length in a
591 sample, which can be calculated as
593 l_bar = \sum_i TPM_i / 10^6 * effective_length_i (i goes through every transcript),
595 the following equation is hold:
597 FPKM_i = 10^3 / l_bar * TPM_i.
599 We can see that the sum of FPKM is not a constant across samples.
601 'IsoPct' stands for isoform percentage. It is the percentage of this
602 transcript's abandunce over its parent gene's abandunce. If its parent
603 gene has only one isoform or the gene information is not provided,
604 this field will be set to 100.
606 'pme_expected_count', 'pme_TPM', 'pme_FPKM' are posterior mean
607 estimates calculated by RSEM's Gibbs sampler. 'IsoPct_from_pme_TPM' is
608 the isoform percentage calculated from 'pme_TPM' values.
610 'TPM_ci_lower_bound', 'TPM_ci_upper_bound', 'FPKM_ci_lower_bound' and
611 'FPKM_ci_upper_bound' are lower(l) and upper(u) bounds of 95%
612 credibility intervals for TPM and FPKM values. The bounds are
613 inclusive (i.e. [l, u]).
615 =item B<sample_name.genes.results>
617 File containing gene level expression estimates. The first line
618 contains column names separated by the tab character. The format of
619 each line in the rest of this file is:
621 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]
623 Fields are separated by the tab character. Fields within "[]" are only
624 presented if '--calc-ci' is set.
626 'transcript_id(s)' is a comma-separated list of transcript_ids
627 belonging to this gene. If no gene information is provided, 'gene_id'
628 and 'transcript_id(s)' are identical (the 'transcript_id').
630 A gene's 'length' and 'effective_length' are
631 defined as the weighted average of its transcripts' lengths and
632 effective lengths (weighted by 'IsoPct'). A gene's abundance estimates
633 are just the sum of its transcripts' abundance estimates.
635 =item B<sample_name.transcript.bam, sample_name.transcript.sorted.bam and sample_name.transcript.sorted.bam.bai>
637 Only generated when --no-bam-output is not specified.
639 'sample_name.transcript.bam' is a BAM-formatted file of read
640 alignments in transcript coordinates. The MAPQ field of each alignment
641 is set to min(100, floor(-10 * log10(1.0 - w) + 0.5)), where w is the
642 posterior probability of that alignment being the true mapping of a
643 read. In addition, RSEM pads a new tag ZW:f:value, where value is a
644 single precision floating number representing the posterior
645 probability. Because this file contains all alignment lines produced
646 by bowtie or user-specified aligners, it can also be used as a
647 replacement of the aligner generated BAM/SAM file. For paired-end
648 reads, if one mate has alignments but the other does not, this file
649 marks the alignable mate as "unmappable" (flag bit 0x4) and appends an
650 optional field "Z0:A:!".
652 'sample_name.transcript.sorted.bam' and
653 'sample_name.transcript.sorted.bam.bai' are the sorted BAM file and
654 indices generated by samtools (included in RSEM package).
656 =item B<sample_name.genome.bam, sample_name.genome.sorted.bam and sample_name.genome.sorted.bam.bai>
658 Only generated when --no-bam-output is not specified and --output-genome-bam is specified.
660 'sample_name.genome.bam' is a BAM-formatted file of read alignments in
661 genomic coordinates. Alignments of reads that have identical genomic
662 coordinates (i.e., alignments to different isoforms that share the
663 same genomic region) are collapsed into one alignment. The MAPQ field
664 of each alignment is set to min(100, floor(-10 * log10(1.0 - w) +
665 0.5)), where w is the posterior probability of that alignment being
666 the true mapping of a read. In addition, RSEM pads a new tag
667 ZW:f:value, where value is a single precision floating number
668 representing the posterior probability. If an alignment is spliced, a
669 XS:A:value tag is also added, where value is either '+' or '-'
670 indicating the strand of the transcript it aligns to.
672 'sample_name.genome.sorted.bam' and 'sample_name.genome.sorted.bam.bai' are the
673 sorted BAM file and indices generated by samtools (included in RSEM package).
675 =item B<sample_name.time>
677 Only generated when --time is specified.
679 It contains time (in seconds) consumed by aligning reads, estimating expression levels and calculating credibility intervals.
681 =item B<sample_name.stat>
683 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.
689 Assume the path to the bowtie executables is in the user's PATH environment variable. Reference files are under '/ref' with name 'mouse_125'.
691 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:
693 rsem-calculate-expression --phred64-quals \
695 --output-genome-bam \
700 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:
702 rsem-calculate-expression -p 8 \
707 mmliver_paired_end_quals
709 3) '/data/mmliver.fa', single-end reads without quality scores. We want to use 8 threads:
711 rsem-calculate-expression -p 8 \
715 mmliver_single_without_quals
717 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:
719 rsem-calculate-expression --bowtie-path /sw/bowtie \
721 --fragment-length-mean 150.0 \
722 --fragment-length-sd 35.0 \
724 --output-genome-bam \
731 5) '/data/mmliver_paired_end_quals.bam', paired-end reads with quality scores. We want to use 8 threads:
733 rsem-calculate-expression --paired-end \
736 /data/mmliver_paired_end_quals.bam \
738 mmliver_paired_end_quals