#const
my $BURNIN = 200;
-my $NSAMPLES = 1000;
+my $CHAINLEN = 1000;
my $SAMPLEGAP = 1;
my $CONFIDENCE = 0.95;
my $NSPC = 50;
my $NMB = 1024; # default
-my $status;
+my $status = 0;
my $read_type = 1; # default, single end with qual
my $nThreads = 1;
my $genBamF = 0;
+my $sampling = 0;
my $calcCI = 0;
my $quiet = 0;
my $help = 0;
my $strand_specific = 0;
+my $mTime = 0;
+my ($time_start, $time_end, $time_alignment, $time_rsem, $time_ci) = (0, 0, 0, 0, 0);
+
GetOptions("keep-intermediate-files" => \$keep_intermediate_files,
"no-qualities" => \$no_qual,
"paired-end" => \$paired_end,
"num-rspd-bins=i" => \$B,
"p|num-threads=i" => \$nThreads,
"out-bam" => \$genBamF,
+ "sampling-for-bam" => \$sampling,
"calc-ci" => \$calcCI,
"ci-memory=i" => \$NMB,
+ "time" => \$mTime,
"q|quiet" => \$quiet,
"h|help" => \$help) or pod2usage(-exitval => 2, -verbose => 2);
pod2usage(-msg => "Min fragment length should be smaller or equal to max fragment length!", -exitval => 2, -verbose => 2) if ($minL > $maxL);
pod2usage(-msg => "The memory allocated for calculating credibility intervals should be at least 1 MB!\n", -exitval => 2, -verbose => 2) if ($NMB < 1);
pod2usage(-msg => "Number of threads should be at least 1!\n", -exitval => 2, -verbose => 2) if ($nThreads < 1);
+pod2usage(-msg => "Seed length should be at least 25!\n", -exitval => 2, -verbose => 2) if ($L < 25);
+pod2usage(-msg => "--sampling-for-bam cannot be specified if --out-bam is not specified!\n", -exitval => 2, -verbose => 2) if ($sampling && !$genBamF);
if ($strand_specific) { $probF = 1.0; }
$command .= " | gzip > $imdName.sam.gz";
print "$command\n";
+
+ if ($mTime) { $time_start = time(); }
+
$status = system($command);
+
+ if ($mTime) { $time_end = time(); $time_alignment = $time_end - $time_start; }
+
if ($status != 0) {
print "bowtie failed! Please check if you provide correct parameters/options for the pipeline!\n";
exit(-1);
$is_sam = 1; # output of bowtie is a sam file
}
+if ($mTime) { $time_start = time(); }
+
$command = $dir."rsem-parse-alignments $refName $sampleName $sampleToken";
my $samInpType;
}
print "\n";
-$status = open(OUTPUT, ">$imdName.mparams");
-if ($status == 0) { print "Cannot generate $imdName.mparams!\n"; exit(-1); }
+my $doesOpen = open(OUTPUT, ">$imdName.mparams");
+if ($doesOpen == 0) { print "Cannot generate $imdName.mparams!\n"; exit(-1); }
print OUTPUT "$minL $maxL\n";
print OUTPUT "$probF\n";
print OUTPUT "$estRSPD\n";
$command .= " -b $samInpType $inpF";
if ($fn_list ne "") { $command .= " 1 $fn_list"; }
else { $command .= " 0"; }
+ if ($sampling) { $command .= " --sampling"; }
}
if ($calcCI) { $command .= " --gibbs-out"; }
if ($quiet) { $command .= " -q"; }
&collectResults("$imdName.iso_res", "$sampleName.isoforms.results"); # isoform level
&collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
+if ($mTime) { $time_end = time(); $time_rsem = $time_end - $time_start; }
+
+if ($mTime) { $time_start = time(); }
+
if ($calcCI) {
- $command = $dir."rsem-run-gibbs $refName $sampleName $sampleToken $BURNIN $NSAMPLES $SAMPLEGAP";
+ $command = $dir."rsem-run-gibbs $refName $sampleName $sampleToken $BURNIN $CHAINLEN $SAMPLEGAP";
# $command .= " -p $nThreads";
if ($quiet) { $command .= " -q"; }
print "$command\n";
&collectResults("$imdName.gene_res", "$sampleName.genes.results"); # gene level
}
+if ($mTime) { $time_end = time(); $time_ci = $time_end - $time_start; }
+
+if ($mTime) { $time_start = time(); }
+
if (!$keep_intermediate_files) {
- $status = system ("rm -rf $temp_dir");
+ $status = system("rm -rf $temp_dir");
if ($status != 0) {
print "Fail to delete the temporary folder!\n";
exit(-1);
}
}
+if ($mTime) { $time_end = time(); }
+
+if ($mTime) {
+ open(OUTPUT, ">$sampleName.time");
+ print OUTPUT "Alignment: $time_alignment s.\n";
+ print OUTPUT "RSEM: $time_rsem s.\n";
+ print OUTPUT "CI: $time_ci s.\n";
+ my $time_del = $time_end - $time_start;
+ print OUTPUT "Delete: $time_del s.\n";
+ close(OUTPUT);
+}
+
# inpF, outF
sub collectResults {
my $local_status;
Generate a BAM file, 'sample_name.bam', with alignments mapped to genomic coordinates and annotated with their posterior probabilities. In addition, RSEM will call samtools (included in RSEM package) to sort and index the bam file. 'sample_name.sorted.bam' and 'sample_name.sorted.bam.bai' will be generated. (Default: off)
+=item B<--sampling-for-bam>
+
+When RSEM generates a BAM file, instead of outputing all alignments a read has with their posterior probabilities, one alignment is sampled and outputed according to the posterior probabilities. If the sampling result is that the read comes from the "noise" transcript, nothing is outputed. It cannot be specified unless --out-bam is specified. (Default: off)
+
=item B<--calc-ci>
Calculate 95% credibility intervals and posterior mean estimates. (Default: off)
=item B<--seed-length> <int>
-Seed length used by the read aligner. Providing the correct value for this parameter is important for RSEM's accuracy if the data are single-end reads. If RSEM runs Bowtie, it uses this value for Bowtie's seed length parameter. (Default: 25)
+Seed length used by the read aligner. Providing the correct value for this parameter is important for RSEM's accuracy if the data are single-end reads. If RSEM runs Bowtie, it uses this value for Bowtie's seed length parameter. The minimum value is 25. Any read with its or at least one of its mates' (for paired-end reads) length less than 25 will be ignored. (Default: 25)
=item B<--tag> <string>
genomic coordinates. Alignments of reads that have identical genomic
coordinates (i.e., alignments to different isoforms that share the
same genomic region) are collapsed into one alignment. The MAPQ field
-of each alignment is set to max(100, floor(-10 * log10(1.0 - w) +
+of each alignment is set to min(100, floor(-10 * log10(1.0 - w) +
0.5)), where w is the posterior probability of that alignment being
the true mapping of a read. In addition, RSEM pads a new tag
ZW:f:value, where value is a single precision floating number
mmliver_paired_end_quals
=cut
+
projects / rsem.git / blobdiff