[EMBOSS] sequence retrieval

[Sean MacEachern]

Hi Jay, 

Just wondering if you have considered the tools from NCBI. If you were to
dload the blast bundle, I think blast-2.2.17 is the most current release,
you can use formatdb to create a blastable database of your fasta seqs that
you can use for blasting using one of the blast programs or retrieving using
fastacmd.

I'm not sure what emboss application you are attempting to use but you could
probably use a for loop to automate some procedure

Eg.

For i in `cat seqIDs.txt`; do fastacmd -d blastdb -s $i > seq.fsa | primer3
-input seq.fsa -output $i_out.primers

Depending on what you want to do something like that might work for you...

Cheers,
Sean


On 6/10/08 4:15 PM, "Jay" <jboddu at uiuc.edu> wrote:

> Daniel:
> I tried seqret in different ways.
> My problem is EMBOSS is not recognizing my master sequence file (which is in
> fasta form) as my private database. Even after I did the indexing using
> dbifasta.
> When seqret is asking me to input sequence(s), I am not able to figure out
> what exactly it accepts.
> I tried dbname:ID, dbname:@listfile.
> I also tried a crude way of copy pasting my master file and listfile in
> "embl" folder in EMBOSSwin folder and try the same syntax (embl:ID,
> embl:@listfile etc.
> These did not work.
> I am assuming that my master file is not being recognized as a private DB.
> I wanted to define my database in .embossrc file. I could not figure this
> out either.
> Jay
> 
> -----Original Message-----
> From: Daniel Barker [mailto:db60 at st-andrews.ac.uk]
> Sent: Tuesday, June 10, 2008 1:50 PM
> To: rls at ebi.ac.uk
> Cc: Peter Rice; Jay; emboss at lists.open-bio.org
> Subject: Re: [EMBOSS] sequence retrieval
> 
> Dear Jay,
> 
> Are you simply trying to extract specific sequences from a Fasta-format
> file? The EMBOSS program to do it is seqret, or maybe seqretsplit:
> 
> http://emboss.sourceforge.net/apps/release/5.0/emboss/apps/seqret.html
> 
> http://emboss.sourceforge.net/apps/release/5.0/emboss/apps/seqretsplit.html
> 
> As Peter Rice suggests, you can do stuff to speed the access up, but
> it'll work without that.
> 
> Best regards,
> 
> Daniel