Skip Menu |

This queue is for tickets about the Bio-SamTools CPAN distribution.

Report information
The Basics
Id: 49986
Status: open
Priority: 0/
Queue: Bio-SamTools

People
Owner: LDS [...] cpan.org
Requestors: nhansen [...] mail.nih.gov
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: (no value)
Fixed in: (no value)



Subject: seq_ids returns only lower-case entry names
Date: Thu, 24 Sep 2009 15:41:48 -0400
To: bug-Bio-SamTools [...] rt.cpan.org
From: Nancy Hansen <nhansen [...] mail.nih.gov>
Hi Lincoln (if this e-mail does indeed get to you!) In your Bio::DB::Sam module, a call to the seq_ids method returns only lower-cased versions of the reference sequences, which is not mentioned in the documentation. I'd rather see it return the exact strings found in the BAM file and/or fasta reference, as I'd like to do something like: my @targets = $sam_obj->seq_ids(); foreach my $this_target (@targets) { my $target_length = $sam_obj->length($this_target); $sam_obj->fasta_pileup("$this_target:1-$target_length", $my_routine, 0); } etc. So I missed chrX inadvertently when I assumed the targets were returned as they were in the BAM file. I can fix this in my own code, but I'd love to see it changed in the next release of the modules. Also, while I have your attention, the code seems to be sucking up memory as I do a pileup through the chromosome, so for large chromosomes, I kludge it by streaming through them in blocks. Just wanted to be sure you were aware... The modules have been a huge help to me. Thanks for releasing them! --Nancy -- ************************************* Nancy F. Hansen, PhD nhansen@nhgri.nih.gov Comparative Genomics Unit, NHGRI 5625 Fishers Lane Rockville, MD 20852 Phone: (301) 435-1560 Fax: (301) 435-6170
You're right: reference Ids should not be converted into lower case. However, you should still be able to designate regions in a case insensitive way, so I will see what I can do to achieve this. Sorry to hear about the memory leak; I thought I had quashed all the leaks. Can you confirm that you are seeing the leak on version 1.04?