BLAST Help

[BioTalk]

Hello all,

I am trying to compare two .fasta files with many fasta sequences within both the files using Blast. But for some reason Blast is considering only first sequence from the fasta files. I am not sure what parameter I should use (for standalone BLAST) to compare all the sequences of both the files.

Please let me know if anyone knows how to do it.

Thank you!

[maubp]

It should "just work".

Which flavour and version of standalone BLAST do you have (the NCBI "legacy" version in C, the new NCBI BLAST+ written in C++, or one of the 3rd party BLAST implementations)?

What BLAST command line are you using?

Have you checked your FASTA files are using the right new line characters for your OS? The command unix2dos or dos2unix can help here. Also try this to count the entries:

grep -c "^>" your_file.fasta

[BioTalk]

Thank you for your prompt response!

I am using Blast for Linux 64 bit downloaded from:

http://blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Web&PAGE_TYPE=BlastDocs&DOC_TYPE=Download

The command line I am using is: blastall -p blastn -F F -W 16 -i <inputfile.fa> -d <knownsequences.fa> -o <outputfile.fa>

[BioTalk]

I am not sure about this question: Have you checked your FASTA files are using the right new line characters for your OS?

[maubp]

I am not sure about this question: Have you checked your FASTA files are using the right new line characters for your OS?

Unix/Linux/Mac OS X etc all use a LF character for a new line, while DOS/Windows uses the CR LF characters. This incompatibility is a common problem when dealing with files created on another OS.

http://en.wikipedia.org/wiki/Newline

[maubp]

Thank you for your prompt response!

I am using Blast for Linux 64 bit downloaded from:

http://blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Web&PAGE_TYPE=BlastDocs&DOC_TYPE=Download

The command line I am using is: blastall -p blastn -F F -W 16 -i <inputfile.fa> -d <knownsequences.fa> -o <outputfile.fa>

If you are using "blastall" then you are using the old legacy BLAST executables based on the NCBI C Toolkit.

If you are using the new BLAST+ suite written in C++ then the command here would be "blastn" instead (and all the options have been renamed).

[BioTalk]

Unix/Linux/Mac OS X etc all use a LF character for a new line, while DOS/Windows uses the CR LF characters. This incompatibility is a common problem when dealing with files created on another OS.
http://en.wikipedia.org/wiki/Newline

Oh okay, I think this is not a problem with the fasta files as they are all created in Linux and being used in Linux. Also, I tried to open the files and they both looks like a normal fasta file.

Please let me know if you know how should I deal with "Blank output file" problem!

[maubp]

Please let me know if you know how should I deal with "Blank output file" problem!

You never mentioned a "blank output file" until now. I thought you said you were having trouble getting BLAST to use multiple input query sequences.


Does blastall give any error messages?

Did you remember to create a BLAST database first using formatdb?

[BioTalk]

I am sorry for the confusion! It is giving almost blank output file with the following details in it instead of alignment result.

BLASTN 2.2.21 [Jun-14-2009]


Reference: Altschul, Stephen F., Thomas L. Madden, Alejandro A. Schaffer,
Jinghui Zhang, Zheng Zhang, Webb Miller, and David J. Lipman (1997),
"Gapped BLAST and PSI-BLAST: a new generation of protein database search
programs", Nucleic Acids Res. 25:3389-3402.

Query= Cluster_573384 1
(23 letters)

So, I thought Blast it not using multiple input query sequences.

[BioTalk]

If you are using "blastall" then you are using the old legacy BLAST executables based on the NCBI C Toolkit.

If you are using the new BLAST+ suite written in C++ then the command here would be "blastp" instead (and all the options have been renamed).

I tried the command for blastn as you have suggested and for that I got some error for indexing.
@biocomp:~/Desktop/Blast/bin$ ./blastn -word_size 16 -query <inputseq.fa> -db <sequencetocompare.fa> -out <outputfile.fa>
BLAST Database error: No alias or index file found for nucleotide database [/home/Desktop/sequencetocompare.fa] in search path [/home/Desktop/Blast/bin::]

[maubp]

You should have one line starting "Query=" for each query sequence.

If that is a full file, it looks like BLAST is crashing or failing to finish.

If I recall correctly, the next output would have been information about the BLAST database - are you sure that is setup right using formatdb? For example, can you do single queries against this database?

[rglover]

It could be that when you've formatted the blast database you didn't set it for nucleotide sequences - formatdb defaults to protein if it finds no command to specify nucleotide.
Try "formatdb -i <yourfasta.fasta> -p F"
The -p F turns protein off and nucleotide on

[BioTalk]

I just tried inputting one query sequence with the command: blastall -p blastn -F F -W 16 -i <inputfile.fa> -d <knownsequences.fa> -o <outputfile.fa>

and I got almost similar output:

BLASTN 2.2.21 [Jun-14-2009]


Reference: Altschul, Stephen F., Thomas L. Madden, Alejandro A. Schaffer,
Jinghui Zhang, Zheng Zhang, Webb Miller, and David J. Lipman (1997),
"Gapped BLAST and PSI-BLAST: a new generation of protein database search
programs", Nucleic Acids Res. 25:3389-3402.

Query= 1-72342
(20 letters)
Do you think there is installation problem? or the command I am using are not correct?

[BioTalk]

It could be that when you've formatted the blast database you didn't set it for nucleotide sequences - formatdb defaults to protein if it finds no command to specify nucleotide.
Try "formatdb -i <yourfasta.fasta> -p F"
The -p F turns protein off and nucleotide on

I tried "formatdb -i <yourfasta.fasta> -p F" and then my previous command but it gave me the same output as before

[rglover]

What are the names of the database files that formatdb is creating? Could you list them here? You could also try putting "-o T" on the end of your formatdb. Other than that I'm not really sure!