hi, all, I used diamond blastx to compare my transcript to the uniprot database, the command is as follows
diamond blastx -d uniprotdb.dmnd -q xxx.fa -e 1e-5 -f 6 -b 10 -o ./xxx.outfmt6 --max-target-seqs 1
The output has 12 columns, which are query id, subject id,% identity, alignment length, mismatches, gap opens, q. start, q. end, s. start, s. end, evalue
I see a lot of gap opens that are not 0, but I do n’t see ‘N’ in my sequence. What ’s the reason? What does the gap opens column mean?
got it, thank you very much.