[Bioperl-l] Restriction::Analysis strange error - please help

Remo Sanges sanges at biogem.it
Wed Jan 26 06:34:20 EST 2005


On Jan 26, 2005, at 3:12 AM, Garrett Sorensen wrote:

> Thanks for the suggestion Rob but still having issues.  I've boiled
> the code down to just reading in fasta sequences and digesting the
> whole sequences - opposed to digesting a subsequence as done
> initially.  It digests a few hundred sequences without issue and then
> runs into the same error, but reports different corrdinates of course.
>
> If it is only reading in a fasta sequence and digesting it how is it
> calculating wrong start/end coordinates for itself?  And the fact that
> it will work great on many sequences but then calculates wrong
> coordinates for one seems strange...  Any ideas?
>
> I've tried feeding Restriction::Analysis both a SeqIO object and a
> PrimarySeq with the same result.
>
> Here is the code:
>
> use strict;
> use Bio::SeqIO;
> use Bio::PrimarySeq;
> use Bio::Restriction::Analysis;
> use Bio::PrimarySeq;
> use Data::Dumper;
>
> my $in  = Bio::SeqIO->new(-file => "$file",  -format => 'fasta');
> while ( my $seq = $in->next_seq ) {
>     my $ra=Bio::Restriction::Analysis->new(-seq=>$seq);
>     my @fragments = $ra->fragments('NlaIII');
>     print join ("\n\n", @fragments);
>     }
> exit
>

Garrett,

this code that you post isn' t really very helpful...
You should have a problem at a certain point in your code
where you ask for a ->subseq with the start bigger than end...
The problem should come from your calculations, not from
module's errors....

HTH

Remo

> On Tue, 25 Jan 2005 17:20:47 -0800, Rob Edwards <rob at salmonella.org> 
> wrote:
>> It is hard to locate the exact error without more information. The
>> error is caused because at some point you are trying to get a sequence
>> that starts at position 2002, but the sequence is only 2001 nt long
>> (hence the error that 2002 must be < 2001). I would suggest that the
>> error is at some point where you are taking the 2kb fragment around 
>> the
>> site. The most obvious thing to start with is what are the start/end
>> coordinates that are called immediately before the error, and do they
>> make sense given the length of the sequence?
>>
>> Rob
>>
>> On Jan 25, 2005, at 9:45 AM, Garrett Sorensen wrote:
>>
>>> Hello,  I'm new to the mailing list.  Thanks in advance for any help.
>>>
>>> I'm really stumped by the following error when running restriction
>>> analysis on large numbers of seq objects.  This only occurs sometimes
>>> when dealing with large numbers of sequences.
>>>
>>>
>>> ------------- EXCEPTION  -------------
>>> MSG: Bad start,end parameters. Start [2002] has to be less than end
>>> [2001]
>>> STACK Bio::PrimarySeq::subseq
>>> /home/garrett/Perl/lib/perl5/site_perl/5.8.5/Bio/PrimarySeq.pm:362
>>> STACK Bio::Seq::subseq
>>> /home/garrett/Perl/lib/perl5/site_perl/5.8.5/Bio/Seq.pm:636
>>> STACK Bio::Restriction::Analysis::fragment_maps
>>> /home/garrett/Perl/lib/perl5/site_perl/5.8.5/Bio/Restriction/
>>> Analysis.pm:552
>>> STACK toplevel Restriction_analyser_multi_CpG_a.pl:182
>>>
>>>
>>> Incase it helps here is what my program is doing:
>>> -Reads in multiple fasta sequences (~7kb average size) on at a time
>>> and creates a SeqIO object for each.
>>> -Restriction sites for a particular enzyme are determined for each
>>> SeqIO object and then a fragment 2kb in size is created around that
>>> site.
>>> -A new Seq object is created using the above fragment using
>>> "$upStreamSeqobj = Bio::Seq->new (-seq => $upStreamSeq);"
>>> -This new Seq object is fed into Restriction Analysis to generate
>>> fragments for another enzyme.
>>>
>>> Thanks so much for any help, best regards,
>>> Garrett
>>> _______________________________________________
>>> Bioperl-l mailing list
>>> Bioperl-l at portal.open-bio.org
>>> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>>>
>>
>>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>



More information about the Bioperl-l mailing list