User:Carl Boettiger/Notebook/Comparative Phylogenetics/2010/05/04

{| width="800"
 * style="background-color: #EEE"|[[Image:owwnotebook_icon.png|128px]] Comparative Phylogenetics
 * style="background-color: #F2F2F2" align="center"|  |Main project page
 * style="background-color: #F2F2F2" align="center"|  |Main project page


 * colspan="2"|
 * colspan="2"|

Wide gap, flat landscape

 * Waiting times on a flatter landscape: sigma_c ~ sigma_k.  Waiting times are dominated by the number of times phase 3 collapses: the established invader in a dimorphic population (right panels).  Collapse of phase 2: collapse of the dimorphism before third population can establish is much rarer (about an order of magnitude).




 * from the run3 dataset, with parameters.

Narrower gap, steeper landscape

 * with sigma_c << sigma_k, branching is easy, neither phase is particularly limiting, though single collapses from both phases are very common.


 * from the run4 dataset.

Slower mutation, bigger steps

 * Trying flat landscape with smaller mu, larger sigma_mu, see if phase 2 becomes limiting. (run5 dataset).  Pars: mu = 5e-4, sigma_mu = .05, sigma_c2 = .8,
 * Just finished running, very few achieved branching! Should repeat with narrower competition kernel.
 * Mean number of failures from phase 2: 240, from phase 3: 330, so not order of magnitude different anymore but phase 3 still limiting.
 * Very interesting, seems the stdev of failures is much smaller for both! (24, 30 respectively)




 * Run6: as before, but mu=.0001, sigma_c2 = .3. Phase 2 is truly more limiting.  A different mechanism to branching -- rare large jump vs series of rapid small steps.




 * }