BLAST Database error: Error pre-fetching sequence data
ramadatta opened this issue · 8 comments
Hi,
I have installed abricate using conda. When I run the command:
$ abricate test.fasta
Using nucl database ncbi: 5386 sequences - 2021-Mar-27
#FILE SEQUENCE START END STRAND GENE COVERAGE COVERAGE_MAP GAPS %COVERAGE %IDENTITY DATABASE ACCESSION PRODUCT RESISTANCE
Processing: test.fasta
BLAST Database error: Error pre-fetching sequence data
Found 0 genes in test.fasta
Tip: it's important to realise abricate uses DNA matching, not AA.
Done.
Can kindly request help on how to resolve this. Many thanks in advance.
Hi
I have same problem with a fresh abricate install with conda.
Also, appreciate any help I can get
Can you confirm the blast version? Thank you.
My blast version is :
blastn: 2.9.0+ Package: blast 2.9.0, build Sep 26 2019 23:31:55
Thank you for your help
The issue comes down to a change in the blast DB format. NCBI updated their default DB format from V4 to V5.
blast 2.9 was the last version that read (and wrote) V4 by default (optionally it would read/write V5). From version 2.11 (I believe) V4 support is dropped all together.
The current blast DB shipped with abricate is V5.
You have two options:
-
Run
abricate --setupdb
which will re-create all the blast DBs. Since you have blast 2.9, it will do them in V4. -
Upgrade to blast v2.10 or above:
conda install -c bioconda blast=2.11
I will work on the conda recipe to update the minimum requirements.
I can't update blast, there are to many conflicts but I succeed with your first solution abricate --setupdb
Thank you very much
I will close this issue. If needed, please re-open.
abricate --setupdb
resolved the issue. Thank you very much!
hey..
Even after updating the blast and also run the command abricate --setupdb my problem is not solved showing the same error blastN database not found...
Also when I am creating the vfdb database it is extracting sequences in duplex, triplex and many more times...
Please help..
Thank you