Which BLAST output format can give full taxonomony information

[SDPA_Pet]

Hi, I was wondering which BLAST output format I should use to do this. I want to blast against nt database. I want the output not only gives me genus and species name, but also gives me phylum, order, family name.

Is it possible?

[jameslz]

you can use the GI number to map the tax id and the get the complete taxonomic lineage information via taxdump.tar.Z. ftp://ftp.ncbi.nlm.nih.gov/pub/taxonomy/

[SDPA_Pet]

Is it possible to map it when I blast it? or I need to get blast result first and then map to tax?

Which command I should use to map the tax id?

[jameslz]

blast output format 6, 7, and 10 can be additionally configured to produce a custom format , it includes the Subject Taxonomy ID (staxids flag),
for example:

blastx -query Human_kinase_rna-100.fasta -db ../ccds/CCDS_protein.20130430 -out Human_kinase-rna-blastx-m7.tbl -evalue 1 -outfmt “7 qseqid qlen slen qcovhsp sseqid staxids bitscore score evalue pident qstart qend sstart send” -num_alignments 10 -num_threads 8

I use a perl script to fetch the complete taxonomic lineage information of the blast staxids.

[SDPA_Pet]

Originally Posted by jameslz

blast output format 6, 7, and 10 can be additionally configured to produce a custom format , it includes the Subject Taxonomy ID (staxids flag),
for example:

I use a perl script to fetch the complete taxonomic lineage information of the blast staxids.

Do I need to put the complete taxonomic lineage information file (ftp://ftp.ncbi.nlm.nih.gov/pub/taxonomy/) in the same folder with my database file? In your example, I didn't see your command call any taxonomic lineage information file.

[SDPA_Pet]

Hi use format 7 and use your command settings

I still do have the taxonomic lineage and I don't know how to use the taxonomic lineage file

Here is my output.

query id query length subject length % hsp coverage subject id subject tax ids bit score score evalue % identity q. start q. end s. start " s. end "
denovo0 266 297 85 gi|14718977|gb|AF352544.1| 163259 351 190 1.00E-93 95.15 39 265 1 219
denovo1 400 1178 100 gi|117572484|gb|DQ979290.1| 1211 706 382 0 98.5 1 400 670 1069

[jameslz]

Usage:

perl tax_trace.pl nodes.dmp names.dmp taxids.txt taxids_export.txt

input file:taxids.txt
format: seqId taxId

gl00001 192
gl00002 2020

nodes.dmp names.dmp can be download from taxdump.tar.Z (ftp://ftp.ncbi.nlm.nih.gov/pub/taxonomy/)


output file:taxids_export.txt

gl00001 Azospirillum brasilense root|cellular organisms|Bacteria|Proteobacteria|Alphaproteobacteria|Rhodospirillales|Rhodospirillaceae|Azospirillum|Azospirillum brasilense no rank|no rank|superkingdom|phylum|class|order|family|genus|species
gl00002 Thermomonospora curvata root|cellular organisms|Bacteria|Actinobacteria|Actinobacteria|Actinobacteridae|Actinomycetales|Streptosporangineae|Thermomonosporaceae|Thermomonospora|Thermomonospora curvata no rank|no rank|superkingdom|phylum|class|subclass|order|suborder|family|genus|species

[SDPA_Pet]

Originally Posted by jameslz

Usage:


input file:taxids.txt
format: seqId taxId


nodes.dmp names.dmp can be download from taxdump.tar.Z (ftp://ftp.ncbi.nlm.nih.gov/pub/taxonomy/)


output file:taxids_export.txt

Thank you. Can you also post your "taxids.txt" and "taxids_export.txt" file. I need to make sure my input file format is right and the output file is what I want. The attachment is my blastoutput and taxid. It didn't work.
Thank you.

[jameslz]

You can parse the blast out file using:

cut -f1,6 blastout.txt >taxids.txt

and then:

perl tax_trace.pl nodes.dmp names.dmp taxids.txt taxids_export.txt

[SDPA_Pet]

Hi, what does this mean? "cut -f1,6 blastout.txt >taxids.txt"

[jameslz]

Originally Posted by SDPA_Pet

Hi, what does this mean? "cut -f1,6 blastout.txt >taxids.txt"

Linux Command, Get the first and the sixth collumn (taxids)

[SDPA_Pet]

Oh, my taxon id is the 5th column. I have another attachment in my post. Is that correct? I can't get taxon information from that file.

[jameslz]

Just fetch the "query id" and "tax ids" collumn .

[maubp]

Which BLAST output format can give full taxonomony information?

Not the XML output (for now), but the optional taxonomy columns in the tabular/comma separated output get you close:
http://blastedbio.blogspot.com/2014/...love-from.html

[yzzhang]

I think blast+ has option to include taxonomy information. the "sscinames" option --- unique Subject Scientific Name(s)

[maubp]

Originally Posted by yzzhang

I think blast+ has option to include taxonomy information. the "sscinames" option --- unique Subject Scientific Name(s)

Yes, but only in the tabular or comma separated output:

• staxids means Subject Taxonomy ID(s), separated by a ';'
• sscinames means Subject Scientific Name(s), separated by a ';'
• scomnames means Subject Common Name(s), separated by a ';'
• sblastnames means Subject Blast Name(s), separated by a ';' (in alphabetical order)
• sskingdoms means Subject Super Kingdom(s), separated by a ';' (in alphabetical order)
• stitle means Subject Title
• salltitles means All Subject Title(s), separated by a '<>'

None of these give the full taxonomy lineage, which is what I think was being asked for here (see the earlier comments).

See also http://blastedbio.blogspot.co.uk/201...criptions.html

[lilicano]

Thanks so much jamslz, your script works great. I dont know but the sskingdoms option had never worked for me.

you made my day with this post!

Best,

Lili

Originally Posted by jameslz

Usage:


input file:taxids.txt
format: seqId taxId


nodes.dmp names.dmp can be download from taxdump.tar.Z (ftp://ftp.ncbi.nlm.nih.gov/pub/taxonomy/)


output file:taxids_export.txt

[sme.bug]

Thanks for sharing! This was very helpful to me.

[maubp]

I believe the fairly new BLAST XML v2 output also includes taxonomy information, if available.

[Brian Bushnell]

Just thought I'd mention JGI's new Taxonomy server. You can look up organism names, gi numbers, or accession numbers to get either the NCBI taxid or complete lineage. These are all in the nt sequence headers.