blastn alighment gives error while excluding sequence from local database by seqidlist
1
0
Entering edit mode
4 weeks ago
rosmakar ▴ 10

Hello Biostar community,

I have several .fa sequences I and I want to check if they have some homology, e.g. if they align within each other. To do so, I concatenate all sequences in a single .fa file, create a database on it, and run blast of the interested .fasta file versus created database, excluding this sequence from db by its Sequence ID. However, blastn returns me an error:

Error: NCBI C++ Exception:
T0 "/build/ncbi-blast+-S1iyIZ/ncbi-blast+-2.9.0/c++/src/corelib/ncbifile.cpp", line 5625: Error: ncbi::CMemoryFileMap::CMemoryFileMap() - To be memory mapped the file must exist: 

My pipeline looks as following:

cat * > fragments_db.fa
makeblastdb -in fragments_db.fa -input_type fasta -dbtype nucl -parse_seqids -out fragments_db
blastn -query PlasmidA.fa -db fragments_db -negative_seqidlist PlasmidA

When exclude parameter -negative_seqidlist from a command above, blastn works as expected, except it do align sequence to itself

My initial .fasta file looks as following and all SeqIDs are unique:

>PlasmidA
atatatgtcaaagaatttagtaaactgtaagacaaaatttaaatgcaagctgttgactat

Could you give me a hint where my problem may come from?

Thank you

blastn blast blastdatabase alignment • 121 views
ADD COMMENT
0
Entering edit mode
4 weeks ago
rosmakar ▴ 10

I found the issue. In fact one needs to provide a fasta file and not just name or its ID. That is what the error about. Therefore the command line should look as following:

blastn -query PlasmidA.fa -db fragments_db -negative_seqidlist PlasmidA.fa

probably there is also a better way to do it

ADD COMMENT

Login before adding your answer.

Traffic: 2744 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6