Perl make-believe vs reality

The computer cluster did finish my job, and I solved an email-notification problem the job had. But in this post I'm going to write about progress with another project, our Perl computer-simulation model of USS evolution.

One of the post-docs and I have been polishing up the Perl simulation, writing a 'narrative' that attempts to explain in plain English what the Perl program does. This is a necessary step to make sure we understand the program correctly before she does a big batch of runs to characterize the equilibria. Now we're asking the other post-doc to help us improve the narrative, so it's sufficiently clear that it makes sense to someone not familiar with the Perl code.

In the course of writing the narrative, we realized that neither of us has a clear understanding of how the steps in the program might correspond to the biological reality we're trying to simulate. Our confusion is not so much about how USS might have been accumulating in the genome over evolutionary time, as about how the events in each cycle of the simulation might correspond to the molecular events of DNA uptake and transformation. So we decided that I should try to explain this in my next lab meeting, which is on Tuesday.

I just made a first pass at explaining it to both postdocs, scribbling all over two of the big whiteboards in the hall. I'm very happy with the result, as we generated the following very clear and accurate statement of what the USS-conversion probabilities in the model really represent. Here it is:

The probability P2 represents, for each USS in the genome (perfect, on-off or two-off USS), the probability that the cell has taken up and recombined a homologous DNA fragment containing a perfect version of this USS. Probability P3 represents the same probability for taking up and recombining a homologous fragment containing a one-off version.