BLAST+ database error

nupurgupta · 06-08-2012, 06:08 AM

There are posts on topics similar to this - but none that exactly address this issue

I have installed BLAST+ on my Linux machine along with a local copy of the microbial database. "blastn" works fine with queries like

blastn -query query.txt -db 16SMicrobial
However, blastp fails e.g.

blastp -query query_protein.txt -db 16SMicrobial
gives

BLAST Database error: No alias or index file found for protein database [16SMicrobial] in search path [..../databases]

My database directory has these files-

16SMicrobial.nhr 16SMicrobial.nnd 16SMicrobial.nog 16SMicrobial.nsi 16SMicrobial.nin 16SMicrobial.nni 16SMicrobial.nsd 16SMicrobial.nsq

My .ncbirc file has

[NCBI]
DATA=.../databases

[BLAST]
BLASTDB=.../databases

Any idea of what could be wrong please?

maubp · 06-08-2012, 06:42 AM

You have a nucleotide database, 16SMicrobial - the clue is the file names are 16SMicrobial.n*

If you are using BLASTP, that is for protein vs protein, and needs a protein database.

If you want to compare your protein file against a nucleotide database, use BLASTX.

Note that you can have both a nucleotide database and a protein database with the same name (e.g. my_database.n* and my_database.p* would both be called my_database).

nupurgupta · 06-08-2012, 06:46 AM

Thank you so much! I just figured it out. But where do I get a protein database from? I'm assuming 16SMicrobial.pal and the like?
Thanks so much

maubp · 06-08-2012, 06:53 AM

In general you can download several nucleotide and protein databases from the NCBI, or make your own from a FASTA file using the makeblastdb command.

In this specific case, it doesn't make biological sense to talk about a protein database for 16S ribosomal RNA sequences, does it?
http://en.wikipedia.org/wiki/16S_ribosomal_RNA

nupurgupta · 06-08-2012, 06:58 AM

Thank you sooo much. Well, I do need to identify the genes in certain parts of the bacterial genomes, so I will go find a fasta file!
Thanks again

GenoMax · 06-08-2012, 07:06 AM

ftp://ftp.jgi-psf.org/pub/IMG/

http://img.jgi.doe.gov/cgi-bin/w/main.cgi

Quote:

Originally Posted by nupurgupta

Thank you sooo much. Well, I do need to identify the genes in certain parts of the bacterial genomes, so I will go find a fasta file!
Thanks again

pag · 06-08-2012, 08:01 AM

If all you're working with is 16S sequences, as mentioned above, you don't want to do protein-based queries.

One database for 16S data is the ribosomal database project: http://rdp.cme.msu.edu/

If you want to use more than just 16S data, why not use the NCBI blast? Are you trying to do automatic annotation based on sequence homology? e.g. this sequence is 80% identical to a sodium channel, so it's probably a sodium channel?

nupurgupta · 06-08-2012, 08:16 AM

OK...no I need to do it against the microbial database. I need to run BLAST on my local machine from the command line. Setting up an automation pipeline. Unfortunately I have proxy issues connecting to the outside , so I need to use a local install of a microbial database. Trying to hunt one down now...(or at least a fasta file)

Thanks soo much!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
BLAST+ creating custom blast database and using blast+ filtering features	deniz	Bioinformatics	2	10-26-2012 11:04 AM
Blast database creation : Error	kalyankpy	Bioinformatics	5	06-18-2012 02:49 PM
Help with Blast and nr database	Avila	Bioinformatics	1	05-16-2012 04:05 AM
BLAST database error - when changing to new BLAST+ local program	biobio	Bioinformatics	4	06-15-2011 05:20 AM
Database of BLAST	CarlElit	Bioinformatics	1	01-04-2010 06:23 AM

06-08-2012, 06:08 AM	#1
nupurgupta Member Location: New Jersey Join Date: Aug 2010 Posts: 28	BLAST+ database error There are posts on topics similar to this - but none that exactly address this issue I have installed BLAST+ on my Linux machine along with a local copy of the microbial database. "blastn" works fine with queries like blastn -query query.txt -db 16SMicrobial However, blastp fails e.g. blastp -query query_protein.txt -db 16SMicrobial gives BLAST Database error: No alias or index file found for protein database [16SMicrobial] in search path [..../databases] My database directory has these files- 16SMicrobial.nhr 16SMicrobial.nnd 16SMicrobial.nog 16SMicrobial.nsi 16SMicrobial.nin 16SMicrobial.nni 16SMicrobial.nsd 16SMicrobial.nsq My .ncbirc file has [NCBI] DATA=.../databases [BLAST] BLASTDB=.../databases Any idea of what could be wrong please?

06-08-2012, 06:42 AM	#2
maubp Peter (Biopython etc) Location: Dundee, Scotland, UK Join Date: Jul 2009 Posts: 1,170	You have a nucleotide database, 16SMicrobial - the clue is the file names are 16SMicrobial.n* If you are using BLASTP, that is for protein vs protein, and needs a protein database. If you want to compare your protein file against a nucleotide database, use BLASTX. Note that you can have both a nucleotide database and a protein database with the same name (e.g. my_database.n* and my_database.p* would both be called my_database).

06-08-2012, 06:46 AM	#3
nupurgupta Member Location: New Jersey Join Date: Aug 2010 Posts: 28	Thank you so much! I just figured it out. But where do I get a protein database from? I'm assuming 16SMicrobial.pal and the like? Thanks so much

06-08-2012, 06:53 AM	#4
maubp Peter (Biopython etc) Location: Dundee, Scotland, UK Join Date: Jul 2009 Posts: 1,170	In general you can download several nucleotide and protein databases from the NCBI, or make your own from a FASTA file using the makeblastdb command. In this specific case, it doesn't make biological sense to talk about a protein database for 16S ribosomal RNA sequences, does it? http://en.wikipedia.org/wiki/16S_ribosomal_RNA

06-08-2012, 06:58 AM	#5
nupurgupta Member Location: New Jersey Join Date: Aug 2010 Posts: 28	Thank you sooo much. Well, I do need to identify the genes in certain parts of the bacterial genomes, so I will go find a fasta file! Thanks again

06-08-2012, 08:01 AM	#7
pag Member Location: USA Join Date: May 2012 Posts: 44	If all you're working with is 16S sequences, as mentioned above, you don't want to do protein-based queries. One database for 16S data is the ribosomal database project: http://rdp.cme.msu.edu/ If you want to use more than just 16S data, why not use the NCBI blast? Are you trying to do automatic annotation based on sequence homology? e.g. this sequence is 80% identical to a sodium channel, so it's probably a sodium channel?

06-08-2012, 08:16 AM	#8
nupurgupta Member Location: New Jersey Join Date: Aug 2010 Posts: 28	OK...no I need to do it against the microbial database. I need to run BLAST on my local machine from the command line. Setting up an automation pipeline. Unfortunately I have proxy issues connecting to the outside , so I need to use a local install of a microbial database. Trying to hunt one down now...(or at least a fasta file) Thanks soo much!