E-utilities pipe to download different genomes

Question

[Mike]

Hi all,

I want to download FASTA from NCBI based on GCF/A ID. I used this command at first:

esearch -db assembly -query GCF_000023565.1 | elink -target nuccore | efetch -format fasta > {genome_out}

to download genomes, but I noticed the result is "duplicated", meaning I get both >NC_ and >CP_ Complete Genomes.

Then I switched to the command

esearch -db nucleotide -query GCF_000023565.1 | efetch -format fasta > {genome_out}

which worked. (File size is half the original).

Now when I try downloading GCF_000648595 with the new command I get an empty file. When I switch back to the original command I do manage to get a fasta file but it seems "duplicated" again.

From the NCBI site I can see that the new command only works on IDs with "WGS project".

I hope some can shed some light on this, I am hoping for a command that can deal with both GCF_000023565.1 & GCF_000648595 examples, and if not than maybe switching between different commands based on some condition.

I have tried multiple other parameters and none seems to work.

Any help will be highly appreciated.

[Mensur Dlakic]

There was a similar question earlier today.

[Mike]

Thanks, the command in that post is the exact one given to me in the said post by GenoMax. But, his was a NCBI error which occurred to me as well but is not the current issue I'm facing.

Answer 1

[GenoMax]

This will get you RefSeq entries.

$ esearch -db assembly -query GCF_000023565.1 | elink -target nuccore | efetch -format docsum | xtract -pattern DocumentSummary -if SourceDb -contains refseq -element Caption | efetch -db nuccore -format fasta

Beware that there can be multiple entries.

$ esearch -db assembly -query GCF_000648595 | elink -target nuccore | efetch -format docsum | xtract -pattern DocumentSummary -if SourceDb -contains refseq -element Caption 
NZ_ADOU02000013
NZ_ADOU02000012
NZ_ADOU02000011
NZ_ADOU02000010
NZ_ADOU02000009
NZ_ADOU02000008
NZ_ADOU02000007
NZ_ADOU02000006
NZ_ADOU02000005
NZ_ADOU02000004
NZ_ADOU02000003
NZ_ADOU02000002
NZ_ADOU02000001

Answer 2

[Mike]

I somewhat solved the issue for now, using the ftp path of each genome to download the zipped fna file and unzipping it.

[GenoMax]

You are better off using NCBI datasets tool for this purpose. EntrezDirect was not designed to do large downloads like the entire genomes.

See --> downloading genomes in fasta format from accession ids

Answer 3

[5heikki]

There's now a more convenient way to do this kind of stuff. Check out NCBI Datasets

[GenoMax]

datasets already recommended in my comment above.