We plan to include example searches in the next release as well as details on analyzing the resulting profiles in an R package we have recently created. In the meantime, simply select one or more target genomes (the columns in the profile table) and optionally include the background, set the number of hits allowed to each group, then choose a cross-reference or cluster identity cutoff and hit search. If you have any additional questions, please send an email to Chris Stubben (stubben at lanl dot gov)