This is a test version of Biostars. For the public version, visit https://www.biostars.org.
can't get staxids with blastn megablast against nt

Hi everyone, Please I really need help. I can't get the staxids with my blastn. This is what i run :

blastn -task megablast -query $RND_ASSEMBLY -db $BLASTDB/nt -evalue 1e-5 -num_threads 5 -max_target_seqs 1 -outfmt "6 qseqid staxids" -out $RND_ASSEMBLY.nt.1e-5.megablast

I downloaded the nt database with the update_blastdb script. In the same directory as my nt database i downloaded the two taxdb file. But despite this i always have this :

299958__len__337 N/A

If I change the parameters in outfmt with salltitles for example it's ok i have a good out file.. what can i do ? thanks so much for your help Marine

blast
• 3,362 views • link
updated by robitaillea • robitaillea • written by marine.bergot • marine.bergot

have you downloaded taxdb? Now you have done the blast, using elink and efetch you can get the taxiD or species name.

• 0 views • link
written by Prasad Prasad

Yes i have downloaded taxdb and the two files are in the same directory as my nt database :/ i don't know elink and efetch i will take a look. But no idea why the command -outfmt "6 qseqid staxids" returned N/A ? thanks

• 0 views • link
written by marine.bergot • marine.bergot

cant think of a reason why NA in your case. here is the elink method to get the species name in a crude form

for i in `awk -F "\t" '{if($1 != a){print $2}a=$1}' blast_file | cut -f 2 -d "|"`; do curl -s "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=protein&db=taxonomy&id=$i&rettype=docsum&retmode=text" | grep -A1 "<Link>" | perl -ne '{chomp; if($_=~/.*Id\>(\d+)/s){print $1,"\n";}}' | while read p; do  curl -s "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=taxonomy&id=${p}&rettype=docsum&retmode=txt" | perl -ne '{if($_=~/\d+.\s+(\w+\s+\w+)/){print '"$i"',"\t", '"$p"',"\t", $1,"\t",$2,"\n";}}'; done; done;
• 0 views • link
written by Prasad Prasad

1 answer

Hi all,

I've got the same problem and I finally find the solution. In the NCBI FTP, there are different level of directory where you can find nt database :

ftp://ftp.ncbi.nlm.nih.gov/blast/db/ & ftp://ftp.ncbi.nlm.nih.gov/blast/db/FASTA/

The second one, in the "FASTA" folder, permit to do a megablast, but does not communicate with taxdb file.

You have to download from the first ftp address (ftp://ftp.ncbi.nlm.nih.gov/blast/db/) where you will find files from "nt.00.tar.gz" to "nt.44.tar.gz". This database allow to retrieve taxonomic information when download the taxdb file in the same directory.

for file in ftp://ftp.ncbi.nlm.nih.gov/blast/db/nt*.tar.gz; do wget ${file}; done
• 0 views • link
written by robitaillea • robitaillea

Log in to answer this question.

More posts like this