[Bioperl-l] Re: Arrays of BioSeq and TCoffee

[James Wasmuth]

Sorry peeps, realised my mistake...

didn't reference the array, kinda misread the documentation, oh and 
didn't read the synopsis at all, wrist has been duely slapped...

I do have a Perl based question tho.  I need to seperate a string into 
an array of fasta sequences.  At the moment I use split command but need 
to give it an expression with which to delimit, the obvious choice is 
'>', but that means I lose it from the header.  I've got round this in a 
clumsy way (couple of lines), but wondered if there was an easy (more 
efficient) way of keeping the '>'?


James Wasmuth wrote:

> Hi people,
>
> I have a number of fasta format sequences in one string, which I wish 
> to align using T-Coffee, without writing the sequences to a file first.
>
> According to doc for T-Coffee module, the align method will take a 
> filename or an array of references for Bio::Seq objects...
>
> I've tried the following script but it fails, I am informed that the 
> first Bio::Seq object contains less than 2 sequences.  Well of course 
> it does, its a Bio::Seq object...  Can anyone see my error...
>
> my @fasta_seq = qw/  >seq1\nFTTATT  >seq2\nFTTGTT  >seq3\nFTATTT /;
> my @seq;
> foreach (@fasta_seq)    {
>        my $stringfh = new IO::String($_);
>        my $seqio = new Bio::SeqIO(-fh => $stringfh, -format => 'fasta');
>        push @seq ($seqio_obj->next_seq);     }
> # here I know that @seq is an array of Bio::Seq objects
>
> my @params = (some params);
> my $factory = new Bio::Tools::Run::Alignment::TCoffee (@params);
> my $aln = $factory->align(@seq);
>
>
> Can a Bio::Seq object hold more than one sequence? Is that my problem
>
> Many Thanks
>
> J
>
>

-- 

Nematode Bioinformatics
Blaxter Nematode Genomics Group
Institute of Cell, Animal and Population Biology
Ashworth Labs
University of Edinburgh
King's Buildings
Edinburgh
EH9 3JT

0131 650 7403