--bowtie-chunkmbs

[rsem.git] / rsem-calculate-expression

diff --git a/rsem-calculate-expression b/rsem-calculate-expression
index 93fabce4686c8d0b7e4fafeae279c6262e7d412e..27243140a19bc25893b75bfb38f83bf6361f028a 100755 (executable)
--- a/rsem-calculate-expression
+++ b/rsem-calculate-expression
@@ -24,6 +24,7 @@ my $C = 2;
 my $E = 99999999;
 my $L = 25;
 my $maxHits = 200;
+my $chunkMbs = 0;      # 0 = use bowtie default
 my $phred33 = 0;
 my $phred64 = 0;
 my $solexa = 0;
@@ -71,6 +72,7 @@ GetOptions("keep-intermediate-files" => \$keep_intermediate_files,
           "bowtie-n=i" => \$C,
           "bowtie-e=i" => \$E,
           "bowtie-m=i" => \$maxHits,
+          "bowtie-chunkmbs=i" => \$chunkMbs,
           "phred33-quals" => \$phred33,
           "phred64-quals" => \$phred64, #solexa1.3-quals" => \$phred64,
           "solexa-quals" => \$solexa,
@@ -110,6 +112,7 @@ pod2usage(-msg => "Min fragment length should be at least 1!", -exitval => 2, -v
 pod2usage(-msg => "Min fragment length should be smaller or equal to max fragment length!", -exitval => 2, -verbose => 2) if ($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);
 
 if ($strand_specific) { $probF = 1.0; }
 
@@ -183,6 +186,8 @@ if (!$is_sam && !$is_bam) {
     $command .= " -p $nThreads -a -m $maxHits -S";
     if ($quiet) { $command .= " --quiet"; }
     
+    $command .= " --chunkmbs $chunkMbs" if $chunkMbs > 0;
+
     $command .= " $refName";
     if ($read_type == 0 || $read_type == 1) {
        $command .= " $mate1_list"; 
@@ -484,7 +489,7 @@ 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>
 
@@ -506,6 +511,10 @@ The path to the bowtie executables. (Default: the path to the bowtie executables
 
 (Bowtie parameter) suppress all alignments for a read if > <int> valid alignments exist. (Default: 200)
 
+=item B<--bowtie-chunkmbs> <int>
+
+(Bowtie parameter) memory allocated for best first alignment calculation (Default: 0 - use bowtie's default)
+
 =item B<--phred33-quals>
 
 Input quality scores are encoded as Phred+33. (Default: on)
@@ -624,7 +633,7 @@ Only generated when --out-bam is specified.
 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
@@ -694,4 +703,3 @@ Assume the path to the bowtie executables is in the user's PATH environment vari
 
 =cut
 
-#  LocalWords:  usr