Extracting all microbial sequences from NT

[kga1978]

Wow, thanks so much guys - this was incredible helpful! I got it all covered now

[Richard Finney]

Easiest method to get taxonomy ids ...
Just check out this directory: ftp://ftp.ncbi.nih.gov/pub/taxonomy/

________________
If you want bacteria and virsus genome in fasta format files ...

Check out doucmentation here :
http://defindit.com/readme_files/ncb...on_format.html
for NCBI file name extensions.

You can ftp download data from NCBI here :
ftp://ftp.ncbi.nlm.nih.gov/genomes/Bacteria/
Look for the all* files. The ftp://ftp.ncbi.nlm.nih.gov/genomes/B...all.fna.tar.gz file should be all bacterial genomes.

Virae here : ftp://ftp.ncbi.nlm.nih.gov/genomes/Viruses/

"WGS bacteria OLD" is thereabouts, just look around. Draft genomes there abouts, too.

_____

Alternate way to get taxon ids for example bacteria ...

You can get the file "all rpt" file via wget :
wget ftp://ftp.ncbi.nlm.nih.gov/genomes/B...all.rpt.tar.gz
Unzip and untar.

Run the command
-bash-3.00$ find . -name '*.rpt' -exec grep Taxid {} \; | sort | uniq
There you go.

[nickloman]

Hey

Not a full solution, but MEGAN provides files which map GIs to taxon IDs for nt and nr via this link: http://ab.inf.uni-tuebingen.de/data/...d/welcome.html

Hope that helps

[kga1978]

Hi all,

I have been trying to find a way to extract all microbial (and eukaryotic) sequences in the NT database but I am running into a bunch of problems.

I have tried to download the GI lists for all bacterial entries using the NCBI nucleotide database, but the generated files always time out and fail to download the file completely. Then I thought maybe I could get the GI IDs using blastdbcmd, but that also fails. I tried the following:

blastdbcmd -db nt -entry all -outfmt '%g %T' | awk '{ if ($2 == "2") print $1 }' > ../gi/bacteria.gi

But that also failed, since the individual entries have their species taxon in the %T field, instead of the domain, etc.

Then I thought maybe I could get a list of all taxon IDs for bacteria, eukaryota, etc., but that also doesn't appear to exist.

So in short - does anybody have an idea how I can extract all microbial sequences (to make a custom database) from the NT database? Whatever method works....

Thanks guys!