Hello All,
I am running blastx with my assembled transcriptome against a local database, which was built from a fasta file. The fasta file contains protein sequences with long accession names. e.g.
>tr|A0A1P8ASE7|A0A1P8ASE7_ARATH Cold-shock protein OS=Arabidopsis thaliana OX=3702 GN=AT1G34049 PE=4 SV=1
After I run blastx with
blastx -db db/prot -query transciptome.fa -out "result.outfmt6" -evalue 1e-20 -outfmt 6 -max_target_seqs 1 -num_threads 64
The result file only contains the abridged accession names. e.g. "evgtrinLocDN2062c1g1t1 A0A1P8ASE7 69.136 81 25 0 597 355 163 243 7.83e-29 109". I want the results contain the full accession names as tr|A0A1P8ASE7|A0A1P8ASE7_ARATH. Is there any setting during the blastx I can change to achieve this?
Many thanks.
2 answers
After contacting the ncbi team, I got the solution. Just in case someone else has the same question.
I forgot to mention that I were using version 2.13.0+ of blast+, the default outfmt6 of which is:
qaccver saccver pident length mismatch gapopen qstart qend sstart send evalue bitscore
Following ncbi team's suggestion, I replace saccver in the default output with sseqid. Now the result format is following:
evgsoapLoc3t2 tr|A0A1P8ASE7|A0A1P8ASE7_ARATH 30.864 162 111 1 126 611 326 486 7.79e-25 97.8
If you use the SequenceServer graphical interface to run your BLAST, it provides a standard table output, and also an extended/full one.
If you download that "Full tabular report" from Sequenceserver, the info you're looking for is in the last 4 columns
Log in to answer this question.
Did you create the database with
-parse_seqidoption?Yes, I did. Will this affect the accession names?