[Bioperl-l] problem with trunc_with_features (SeqUtils.pm)

Roy Chaudhuri roy.chaudhuri at gmail.com
Tue Apr 10 10:45:21 EDT 2012


Turns out I spoke too soon, I added in some new tests and they 
highlighted problems with both trunc_with_features and 
revcom_with_features. I think I have resolved all the issues in the most 
recent Redmine patch - Frank, Heath, please could you check that it 
works for you?

Cheers,
Roy.

On 10/04/2012 12:52, Frank Schwach wrote:
> Brilliant, thanks Roy!
> Frank
>
>
> On 10/04/12 12:10, Roy Chaudhuri wrote:
>> Hi Heath, Frank,
>>
>> This was probably my fault back in the mists of time. Looks like an easy
>> fix though, I've reported the issue on Redmine and submitted a patch:
>> https://redmine.open-bio.org/issues/3339
>>
>> We should probably also add Heath's example as a test case.
>>
>> Cheers,
>> Roy.
>>
>> On 10/04/2012 09:42, Frank Schwach wrote:
>>> Hi Heath,
>>>
>>> Yes, I just had a look too and it's true that it would currently ignore
>>> the original type. I had added some new methods (delete, insert, ligate)
>>> and with those the location type is preserved but not with the already
>>> existing methods like trunc_with_features. I will look into it when I
>>> have some time and make some changes.
>>>
>>> Cheers,
>>>
>>> Frank
>>>
>>>
>>> On 09/04/12 22:37, Heath O'Brien wrote:
>>>> Hi Frank,
>>>>
>>>> I just tried it with the latest version from bioperl-live, and it worked
>>>> the way I described in my email.
>>>>
>>>> all good things,
>>>> Heath
>>>>
>>>>
>>>> On 9-Apr-12, at 5:21 PM, Frank Schwach wrote:
>>>>
>>>>> Hi Heath,
>>>>>
>>>>> I have recently worked a bit on that module and contributed the code
>>>>> to bioperl-live. I think this behaviour may already have changed but
>>>>> I'm not 100% sure at the moment. When I have some time I will review
>>>>> the code to confirm. In the meantime, you could give it a go with the
>>>>> bioperl-live version if that's an option for you?
>>>>>
>>>>> Cheers,
>>>>>
>>>>> Frank
>>>>>
>>>>>
>>>>> On 03/04/12 17:56, Heath O'Brien wrote:
>>>>>> Hi All,
>>>>>>
>>>>>> I've encountered a bug in the trunc_with_features function in
>>>>>> SeqUtils.pm, or at
>>>>>> least behavior that was unexpected to me:
>>>>>>
>>>>>> Features with fuzzy coordinates in the original sequence are
>>>>>> converted to exact
>>>>>> coordinates in the truncated sequence. For example, the script below
>>>>>> changes the
>>>>>> coordinates for the feature from<1..5 to 1..5.
>>>>>>
>>>>>> I have modified the code to change this behavior on my system, but I
>>>>>> thought I'd
>>>>>> post something here in case others encounter the same problem.
>>>>>>
>>>>>> all good things,
>>>>>> Heath
>>>>>>
>>>>>>
>>>>>>
>>>>>> #!/usr/bin/perl -w
>>>>>>
>>>>>> use strict;
>>>>>> use warnings;
>>>>>> use Bio::SeqIO;
>>>>>> use Bio::SeqUtils;
>>>>>>
>>>>>> my $infile= shift;
>>>>>>
>>>>>> my $inIO = Bio::SeqIO->new('-file' =>  $infile,
>>>>>> '-format' =>  'genbank') or die "could not open seq file $infile\n";
>>>>>>
>>>>>> my $outfile = $infile . '_out.gbk';
>>>>>>
>>>>>> my $outIO = Bio::SeqIO->new('-file' =>  ">$outfile",
>>>>>> '-format' =>  'genbank') or die "could not open seq file $outfile\n";
>>>>>>
>>>>>> my $in_seq = $inIO->next_seq;
>>>>>> my $out_seq = Bio::SeqUtils->trunc_with_features($in_seq, 1, 5);
>>>>>> $outIO->write_seq($out_seq);
>>>>>> exit;
>>>>>>
>>>>>>
>>>>>> LOCUS test_sequence 57303 bp DNA linear UNA
>>>>>> DEFINITION Sequence to demonstrate unexpected behavior of
>>>>>> trunc_with_features
>>>>>> ACCESSION unknown
>>>>>> KEYWORDS .
>>>>>> FEATURES Location/Qualifiers
>>>>>> source 1..10
>>>>>> /mol_type="genomic DNA"
>>>>>> gene<1..5
>>>>>> /gene="test"
>>>>>> CDS<1..5
>>>>>> /product="hypothetical protein"
>>>>>> ORIGIN
>>>>>> 1 caagattaaa
>>>>>> //
>>>>>>
>>>>>> _______________________________________________
>>>>>> Bioperl-l mailing list
>>>>>> Bioperl-l at lists.open-bio.org
>>>>>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>>>>>
>>>>>
>>>>> --
>>>>> The Wellcome Trust Sanger Institute is operated by Genome Research
>>>>> Limited, a charity registered in England with number 1021457 and a
>>>>> company registered in England with number 2742969, whose registered
>>>>> office is 215 Euston Road, London, NW1 2BE.
>>>>
>>>
>>>
>>
>
>



More information about the Bioperl-l mailing list