Does anyone know if there is some publicly available ribosomal interval list for use with CollectRnaSeqMetrics in the Picard suite? Specifically I'm looking for a interval list for the Human genome in build g1k_v37. I found a description of how to generate my own in the faq but for obvious reasons it would be nice if there is one ready to use somewhere out there. Googleing hasn't turned anything up for me this far, so I thought I'd turn to the community to ask.
Hi Johan,
I was researching this exact question today and came across this post by Alec Wysoker, the author of Picard tools:
http://sourceforge.net/mailarchive/message.php?msg_id=27560147
I would just post his list here but it exceeds the character limit. I had to take out the @SQ lines referencing "GL" locations (the lines after 26), but other than that it worked great for me.
EDIT: You need to make sure your chromosome names in the @SQ lines match the first column in your interval list. I can give more detail if you don't already know what I'm talking about.
EDIT 2: Also ran across this site while searching for mouse:
I don't know if that list is complete if you're using all the contigs. We get a lot of rRNA reads aligning to contig GL000220.1
Thank you for sharing Kamil. I made a couple of changes to the script use other species and rRNA can be in bed format:
Of course, you have been acknowledged. I hope you don't mind :)
•
written
by
A. Domingues
Using this make_rRNA.sh script, I was able to generate the interval list very easily, however the interval list generated has missing headers, (for example lines starting with @SQ). Any pointers what might be the possible issue here?
imaparna27 Seems that there is something wrong with your line perl -lane 'print "\@SQ\tSN:$F[0]\tLN:$F[1]\tAS:$ENV{'genome'}"' $chrom_sizes | \ grep -v _ \
>> $rRNA
On the other hand, I have run this for GRCh38.p13 and its important to use gene_type and not gene_biotype... and big question here:
is it transcripts or gene the type we must select? the 2 scripts differ on that..
and also, if I select transcripts I get very low amount of transcrips... is that right?
Log in to answer this question.
Is there one for hg38?