Time to stop playing around with the model

I'm in Philly for the American Society for Microbiology meeting (talking in the panel on Open Science tomorrow afternoon). The meeting hasn't started yet, and I've been playing around with my Perl simulation of uptake sequence evolution, trying half-heartedly to figure out why uptake sequences aren't being maintained. But this isn't getting me much of anywhere, and what I really need to do is buckle down and sort out what we learned from all the runs the post-doc did before she moved on.

The playing around was largely motivated by a worry that something was wrong with the program component that gradually reduces the uptake bias when not enough fragments have recombined. The bias seemed to be becoming very low very fast, and this could explain why the model couldn't maintain the uptake sequences in the genome segment I've been giving it. I've now convinced myself that the bias-reduction component is behaving as it should, so the low bias must mean that the program is going through MANY sets of fragments before it finds enough it likes. Which probably means it's recombining the same fragments over and over within a single evolutionary cycle, not at all what I want to happen.