Best method to retrieve genome sequences

Question

[kay.hellboy]

What is the best method to get all sequences of the eukaryotic genomes. (just the DNA sequence of the chromosomes)

I wrote a code in perl to retrieve the sequences by accession. but I don't have all the accessions and ensembl will probably block my IP if try to download them all programmatically.

[Bert Overduin]

I don't think Ensembl / Ensembl Genomes will IP block you if you retrieve genome sequences from their respective ftp sites. IP blocks are in general only given to people who scrape the website and thus bring down the production servers.

[kay.hellboy]

I read it on their website, that repeated requests can be considered as abusive :S so not sure about that.

[Bert Overduin]

I am sure downloading many files from the ftp site is not considered as abusive. Cheers, Bert (Ensembl team member April 2005 - March 2014 :) )

Answer 1

[MAPK]

You can look for taxids, extract all the GI/ accession numbers from those species in NCBI. Once you have GIs, you can download sequences from NCBIs nr/nt database using blastdbcmd -batch_entrez option in standalone blast (don't remember the command exactly).

[kay.hellboy]

I'll give it a try, thank you

[kay.hellboy]

But I wouldn't know which GI is for the chromosome, because it holds all the GIs of that taxid without indication to its type

[MAPK]

I think that is correct, you have to figure out the way to filter out the mitochondrial sequences.

[kay.hellboy]

I found a file one the ncbi genome ftp page that has all the IDs for each kingdom, made it a lot easier. thank you for the hint