Re: [Seqan-dev] reading fasta with non-DNA5 characters

[Hannes Hauswedell <hauswedell@mi.fu-berlin.de>]

Hi,

it all depends on what type of String you give to the recordReader. It 
is designed to check for the SequenceType's alphabet and use that as a 
criterium, so the following should fail on non-ACGTN:


CharString seqIds;
Dna5String faSeqs;

RecordReader<String<char, MMap<> >, DoublePass<Mapped> > 
refReader(seqMMapString);
int read2out = read2(seqIds, faSeqs, refReader, Fasta());


but the following should accept any alphabetical[1]:


CharString seqIds;
CharString faSeqs;

RecordReader<String<char, MMap<> >, DoublePass<Mapped> > 
refReader(seqMMapString);
int read2out = read2(seqIds, faSeqs, refReader, Fasta());


This won't however convert anything. That could be done later, maybe 
using ModifiedString<> to avoid copying.

Regards,
Hannes


[1] I just checked that because I want to read a sequence including 
gaps, which fails (because '-' is non alphabetical --working on 
patch--). However the mentioned characters  Y and M should not be a problem.


On 04/26/2012 06:29 AM, Holtgrewe, Manuel wrote:
> If I remember correctly, the<seqan/stream.h>  interface will not allow you to read non-ACGTN characters when using read(..., Fasta()) but return an error value != 0 in this case.
>
> Bernd, what you can do right now is to load your reads into CharStrings and convert them to Dna5.
>
> In the long term, I guess we will need more control for the users over I/O behaviour. At the moment, the assumptions are fairly conservative, e.g. do not allow non-ACTG(N) characters for Dna(5) and tailored to what you find in NGS reads and whole genome data.
>
> Cheers,
> Manuel
> ________________________________________
> From: Weese, David [weese@campus.fu-berlin.de]
> Sent: Wednesday, April 25, 2012 10:44 PM
> To: SeqAn Development
> Subject: Re: [Seqan-dev] reading fasta with non-DNA5 characters
>
> Hi,
>
> actually it should automatically convert every non-ACGT character to N (or A for Dna targets). Have you already tried reading your files into string over Dna5 alphabets?
>
> Cheers,
> David
> --
> David Weese                             weese@inf.fu-berlin.de
> Freie Universität Berlin                http://www.inf.fu-berlin.de/
> Institut für Informatik                 Phone: +49 30 838 75246
> Takustraße 9                                    Algorithmic Bioinformatics
> 14195 Berlin                                    Room 021
>
> Am 25.04.2012 um 11:08 schrieb Bernd Jagla:
>
> > Hi,
> >
> > I have a couple of genome seqeunces that contain characters other than ACTGN (i.e. Y, M,...)...
> >
> > Is there a way to read those sequences in as well and automatically convert those non conforming letters to N?
> >
> > Thanks,
> >
> > Bernd
> >
> > PS:
> >
> > I am using:
> >
> >         RecordReader<String<char, MMap<>  >, DoublePass<Mapped>  >  refReader(seqMMapString);
> >         int read2out = read2(seqIds, faSeqs, refReader, Fasta());
> >
> > for reading in the data and get an INVALID_FORMAT error...
> >
> >
> > _______________________________________________
> > seqan-dev mailing list
> > seqan-dev@lists.fu-berlin.de
> > https://lists.fu-berlin.de/listinfo/seqan-dev
>
> _______________________________________________
> seqan-dev mailing list
> seqan-dev@lists.fu-berlin.de
> https://lists.fu-berlin.de/listinfo/seqan-dev
>
> _______________________________________________
> seqan-dev mailing list
> seqan-dev@lists.fu-berlin.de
> https://lists.fu-berlin.de/listinfo/seqan-dev