[Bioperl-l] Re: Best method for downloading 100 sequences

James Wasmuth james.wasmuth at ed.ac.uk
Fri Sep 2 14:46:13 EDT 2005

 From past experience the lovely people at EMBL only get cross and hunt 
you down if you do it for thousands of sequences.


Mauricio Herrera Cuadra wrote:

> Hi Amir,
> I've done it before by using a loop without sleep() for the EMBL-Bank 
> and it works fine. I use an input file with all the sequence ID's 
> organized by gene/allele name so all the downloaded files get into 
> corresponding directories. I can send you the code if you want to 
> adjust it to your needs.
> Regards,
> Mauricio.
> --- Amir Karger <akarger at CGR.Harvard.edu> wrote:
> > Hi.
> >
> > I'm using Bioperl's nice get_sequence in my Scriptome toolbox, to
> > fetch a single sequence. What would be the best method for downloading
> > 100 sequences? Do I write a loop to call get_sequence N times? Will
> > the various websites get angry at me for doing that? Would they be
> > less angry if I did a 1-second sleep after each download? I know NCBI
> > has methods to pull in N sequences, but I don't know whether Swiss et
> > al. do too. I'm happy to use other Bioperl code, rather than
> > get_sequence. I just need to have a script that people can cut and
> > paste, where they just input the filename with sequence IDs and the
> > database to download from (sort of like
> > 
> http://cgr.harvard.edu/cbg/scriptome/Tools/Fetch.html#fetch_a_sequence_from_ 
> > a_popular_internet_database__fetch_sequence_web_)
> >
> > Thanks,
> >
> > -Amir Karger

"The best model of a cat is another cat or,
	 better, the cat itself"
		-Norbert Wiener

Blaxter Nematode Genomics Group   |
Institute of Evolutionary Biology |
Ashworth Laboratories, KB         | tel: +44 131 650 7403
University of Edinburgh           | web: www.nematodes.org/~james
Edinburgh                         |
EH9 3JT                           |
UK                                |

More information about the Bioperl-l mailing list