makeblastdb error, but not when run from commandline

Question: makeblastdb error, but not when run from commandline

3.5 years ago by

United States

danielfortin86 • 110 wrote:

Hi,

I'm trying to create at blastDB and get errors. When I run the command that Galaxy uses at the command line it finishes with no errors. I'm not understanding what the issue is. Can anyone shed some light on this?

Fatal error: Matched on Error:
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given
Error: (1431.1) FASTA-Reader: Warning: FASTA-Reader: No residues given

ncbi blast makeblastdb • 2.3k views

ADD COMMENT • link •

modified 3.5 years ago by Jennifer Hillman Jackson ♦ 25k • written 3.5 years ago by danielfortin86 • 110

3.5 years ago by

Jennifer Hillman Jackson ♦ 25k

United States

Jennifer Hillman Jackson ♦ 25k wrote:

Hello,

This is setting up blastdb databases for megablast, for use in Galaxy? On the command-line? This is not really a Galaxy issue.

That said, help for creating indexes is in the galaxyproject.org wiki: http://wiki.galaxyproject.org/Admin/DataPreparation

I would just use the regular command line (unless you can simply pull the indexes directly from NCBI, much better IMHO, when available). If it errors, like above, double check the input is really in strict fasta and that the options used are appropriate for the reference genome (protein vs nucleotide). Test the indexes with a simple self-hit line-command. Then put indexes into your data area and update the blastdb.loc.

If there really is a problem with the tool, then that is a different case. I haven't seen anything online about there being a problem with this function, but you never know, corner cases exist. Maybe search as well?

If this is not what you are asking, please clarify and we can try to help. Jen, Galaxy team

ADD COMMENT • link written 3.5 years ago by Jennifer Hillman Jackson ♦ 25k

One last thought .. are you using a protein database with the default Galaxy Megablast wrapper? In Galaxy, as a job? I believe that tool has default configured to work with nucleotide DBs. SInce I haven't attempted to use it with a protein database, the error is not one I recognize, but perhaps this helps anyway? Jen

ADD REPLY • link written 3.5 years ago by Jennifer Hillman Jackson ♦ 25k

Sorry, my post wasn't as clear as it could be. The FASTA file was generated using the Galaxy Tool "Get FASTA from Taxon" and I'm trying to create a BLASTDB using the Galaxy wrapper for BLAST (NCBI BLAST+ makeblastdb). This is so that users can create their own custom BLASTDBs from specific taxons. When I use this wrapper I got the errors mentioned above. But when I cut and paste the command that Galaxy is using onto the command line I get no errors returned and the database is created successfully. Unless I'm missing something this must mean that the error is either in Galaxy or in the makeblastdb wrapper.

ADD REPLY • link written 3.5 years ago by danielfortin86 • 110

Hi Daniel

When you execute the command on the command line (without errors) are you running it as the same user the galaxy server is running? Different environment variabbles might explain the different outcomes.

Also, just double checking, are you really using the same fasta file in both cases?

Third, have you looked at Jen's second comment?

Hans-Rudolf

ADD REPLY • link modified 3.5 years ago • written 3.5 years ago by Hotz, Hans-Rudolf • 1.8k

Hi Hans-Rudolf,

Yes, when I run the command it is also as the the same user as the galaxy server is running. I copy and paste the command being used by Galaxy so the files should be identical. I'm not trying to use megablast at all, when I use makeblastdb it's with the nucleotide option, not with protein.

This is the command that Galaxy gives me when I click on the information icon and look at Job Command-line

python /opt/shed_tools/toolshed.g2.bx.psu.edu/repos/devteam/ncbi_blast_plus/2fe07f50a41e/ncbi_blast_plus/tools/ncbi_blast_plus/check_no_duplicates.py "/remote/users/galaxy/files/002/dataset_2928.dat" && makeblastdb -out "/remote/users/galaxy/job_working_directory/001/1308/dataset_3203_files/blastdb" -hash_index -in "/remote/users/galaxy/files/002/dataset_2928.dat " -title "hg" -dbtype nucl > "/remote/users/galaxy/files/003/dataset_3203.dat"

Interestingly, there seems to be a correct blast database at /remote/users/galaxy/job_working_directory/001/1308/dataset_3203_files/blastdb, but I can't use it in galaxy since it is red. Is there a way to force galaxy to change it to green?

ADD REPLY • link written 3.5 years ago by danielfortin86 • 110

Hi Daniel

Well, if the Blast indices are correct, then the history item is red just because of the warning/error messages. Do you get the same warning/error messages when you run it on the command line?

You can force it to be green, BUT THIS IS NOT RECOMMENDED, by changing the value for 'state' in the 'job' table (in the PostgreSQL database) from "error" to "ok" for this particular job. Make sure you have an up-to-date back-up of your database, before you do any attempts to change it.

Hans-Rudolf

ADD REPLY • link written 3.5 years ago by Hotz, Hans-Rudolf • 1.8k

I don't get the same error messages when run from the command-line

This is a separate issue, but I just tried the same process with a different fasta file input. This new input finishes correctly (green), but I cannot select this database as a database from user history within the ncbi blast+ blastn tool. Additionally, when I go to the directory where the BLAST DB indexes should be located (/remote/users/galaxy/job_working_directory/001/1310/dataset_3205_files/blastdb) they are not there. Is there something wrong with my blast+ installation? The databases do remain if I change "cleanup_job=always" to "cleanup_job=never" but this shouldn't be necessary should it?

Galaxy Job Command-line:

python /opt/shed_tools/toolshed.g2.bx.psu.edu/repos/devteam/ncbi_blast_plus/2fe07f50a41e/ncbi_blast_plus/tools/ncbi_blast_plus/check_no_duplicates.py "/remote/users/galaxy/upload/Genomes/TG3.fa" && makeblastdb -out "/remote/users/galaxy/job_working_directory/001/1310/dataset_3205_files/blastdb" -hash_index -in "/remote/users/galaxy/upload/Genomes/TG3.fa " -title "xsf" -dbtype nucl > "/remote/users/galaxy/files/003/dataset_3205.dat"

ADD REPLY • link modified 3.5 years ago • written 3.5 years ago by danielfortin86 • 110

"I don't get the same error messages when run from the command-line"

this would tell me, although run by the same user, that the environemt is different.

"This is a separate issue, but I just tried the same process with a different fasta file input. This new input finishes correctly (green)"

....maybe Peter C. and or Bjoern G can help here?

ADD REPLY • link written 3.5 years ago by Hotz, Hans-Rudolf • 1.8k

Similar posts • Search »