Reading: Stadler, T. (2010). Sampling-through-time in birth–death trees. Journal of Theoretical Biology, 267(3), 396-404. https://doi.org/10.1016/j.jtbi.2010.09.010.
This paper develops the foundations of the birth-death approach to phylodynamic inference which is the main current alternative to methods based on extensions Kingman’s coalescent. You are expected to spend at least 2hr studying this paper.
The arguments depend heavily on an independent self-similarity property of the trees: any two lineages extant at time t give rise to independent and identically distributed birth-death trees.
If you read through to Theorem 3.5 you are doing well. This result gives the likelihood function for a birth-death phylodynamic model. At some point, you may jump to Sections 4 and 5 to get some additional context.
Beyond Theorem 3.5, the additional results focus on different conditioning events. You might like to think about why these additional results are of scienfific interest.
Class 3 had a lively and productive discussion, but there were some missing voices. The intention is for everyone to get heard. With this in mind, we will offer an alternative to speaking up in class. If you wish, you can get your class participation credit by submitting ahead of class, in writing, a few sentences describing (i) something you got out of reading this paper; (ii) something you got stuck on trying to understand. You can email this to ionides@umich.edu and kingaa@umich.edu with subject “700 participation”.
Note that each class is graded with one point for attendance and one for participation.