George suggested the following, which is a neat idea:
As a feature, could you have a ring or maybe color the reference by the levels of sequence conservation, summaring all BLAST rings.... Say, if there were 20 samples shown, some sites in the genome may be absent/variable in only one sample, while others may often be absent/variable