Hi, Phil! > Is there a way I can force blast to use the full 347 amino acids for > comparison. No. It does not make sense to put sequences into an alignment that do not belong together. > The researchers in my lab are most interested in places > with low similarities, since they are trying to find the portions > which make this organism virulent. Blast can create output in form of a table. Or you can choose the output option XML and convert that into a table. You can filter this table for genes which made only a bad hit. Where "bad" would have to be defined carefully. Then determine which genes did not make a blast hit at all. That way you would get a set of candidate genes causing virulence. (As far as this is possible with this kind of analysis.) Greetings from Germany, Michael.