[ssml] Small Genome Similarity

[Kevin Karplus]

I'm not an expert in searching for homologs of DNA sequences, but if I
had to do it, I'd probably try a blastn or tblastn against all of
genback, and a blastx against nr.

If there were any hits, I might try better alignment methods to try to
find the whole genes and correct for any frameshifts that might have
occurred in sequencing (or due to wierd translational regulation).

If there are no hits with blast on the whole genome, you could try
splitting it up into orfs and trying the orfs separately, but this is
unlikely to help.

If you are sure enough of the sequencing, you could try translating
each ORF and submitting the hypotheical gene product to psi-blast or
to a protein-structure prediction server.

Kevin Karplus