16 March 2012

Bodega Day??? - Its all a blur

Third night in a row that I have been hunched over the computers at 2:00 am.  I have built trees, evolved sequences, inferred trees, set priors, I have run Beast, Mr Bayes, Beuti, Tracer, Seqgen, and 20 other programs plus a lot of scripts.  I have assessed convergence. I have two computers running different simulated dataset through MrBayes right this minute......and I have .t files that look like pin cushions!!!!!  I am pretty sure that most of that is due to the kind of missing data sets that I am simulating.  but.......maybe its my priors.......    Bodega is really intense and I feel really drained but I have learned more this week about phylogenetic methods than I have learned in the last year and a half.  And even though I am tired I am inspired to go home and do some real work in Bayesian framework.  When this topic is taught well it makes so much sense and you can see the importance of making our estimates and inferences across a posterior distribution of trees.  I was confused before coming here thinking that somehow a Bayesian approach was prone to bias now I feel that a non Bayesian approach is prone to bias due to an unrealistic specification of parameters.  I get to give a short presentation tomorrow which should be fun.  Here is a pretty picture I took just down the road from Bodega.  This is a member of the famous ring species complex in the genus Ensatina.


No comments:

Post a Comment