Discuss evolution reading group

Summary of Tangled Nature discussion

The discussion centred around Christensen et al. and Schwämmle & Brigatti. One problem we identified was that there were no "methods" sections: you have to follow all through the papers to understand what exactly is being simulated. It would have been nice to know, for example, how exactly the linear regressions were done in Figure 2 of Schwämmle & Brigatti; as it is they seem to have been rather cavalierly drawn in, although the error bounds seem to indicate that a statistical method has been employed.

John was particularly happy to see that a non-asexual model was introduced in Christensen et al., since so much theory gets done on asexuals. At the same time though it was unsatisfactory that the models are so limited, but there has to be a balance struck between sophistication and tractability. Finally we found Figure 1 of Schwämmle & Brigatti fascinating since it demonstrated some species reuniting after some time evolving separately; although it seems to me this would be considerably less likely to occur if the "strategies" were multiple-dimensional rather than one-dimensional.

posted by John S. Wilkins @ 3:21 PM