I'm using SequenceServer to BLAST RADseq reads against a custom BLAST database constructed from a genome which I've just assembled.
However, I'm getting a lot of hits which I do not want. I only want hits with a perfect match to the cut site of the restriction endonuclease used to make the RAD libraries (EcoRI - G'AATT,C), at the start of the hit.
Is there a way to coerce BLASTn to only return hits with a perfect match to this sequence at the beginning of the hit, but which are free to vary "normally" downstream of that sequence?
All my RADseq reads start GAATTC. (Just in case anyone is wondering - I've added a G to the beginning of all my RADseq reads to ensure that I only get hits which match the whole cut site, but I'm still getting hits which begin at e.g. position 7 of the query sequence).