fork, change, commit, issue pull request - it will be incorporated.
Lincoln wrote the module to support gbrowse in the first iteration - I
think I have been annoyed by the same problems as you but found it was
easier for me to just create a new object most of the time rather than try
to worry about some of the speed things that it does under the hood. I
think adding the warnings is your best bet though rather than retooling it
to allow objects to be mutable.
On Jun 21, 2013, at 5:08 AM, "Freeman, Robert M."
> So, a number of things bother me about this:
> - why is BioPerl deciding that I can't change this?
> - no warning is thrown when I try to change any value of the object
> - documentation does not indicate the object is immutable
> - documentation does not indicate that a PrimarySeq is returned.
> - Worse (at the time), the object is tied to the DB file, so when I
dumped the object in the debugger, it started dumping the whole DB Index.
Not sure why this decision was made.
> LOL! Are these bugs or features?
> Yeah, an option at fetch time would be nice.
> Sent from my iPhone
>> On Jun 20, 2013, at 10:13 PM, "Fields, Christopher J"
>>> On Jun 20, 2013, at 6:27 PM, Bob Freeman
>>> Hoping for an easy answer on this…
>>> Using Bio::DB::Fasta to retrive a seqobj with get_Seq_by_id. I can
print the raw sequence with seqobj->seq(), but when I try to change and set
the raw sequence data (e.g. seqobj->seq("ACTT"), the change is ignored (as
evidenced when writing out to new FASTA file).
>>> Looking under the hood with the Perl debugger, it appears that the
object returned is a Bio::PrimarySeq::FASTA, which points to the whole
(indexed) FASTA file; not what I had expected.
>>> Is the PrimarySeq obj that is returned immutable?
>> Right; the object is a lightweight representation IIRC and is designed
to be immutable, I'm sure b/c the data in the 'database' (file) can't be
>>> If so, is there an easy way to 'copy' the object so that I can change
the sequence and then write this out to a different file?
>> I would copy them to a new Bio::PrimarySeq; it's lighter that a full
Bio::Seq and will work with Bio::SeqIO. Might be worth adding a method to
Bio::DB::Fasta to do this for convenience.
>>> Bob Freeman, Ph.D.
>>> Acorn Worm Informatics, Kirschner lab
>>> Dept of Systems Biology, Alpert 524
>>> Harvard Medical School
>>> 200 Longwood Avenue
>>> Boston, MA 02115
>>> 617/432.2294, vox
>>> "Sorry I'm late. Oh, God, that sounded insincere. I'm late."
>>> -- Karen Walker, from Will and Grace
>>> Bioperl-l mailing list
>>> [email protected]
> Bioperl-l mailing list
> [email protected]