Working notes for Goteborg, 2012

Programming notes for models and simulations

2012-01-09 My implementation of MUL-tree model. Documentation for my code as it develops. Implementation in BEAST of multiply-labelled tree model AlloppMUL. See 2011-09-15 for first version of AlloppNET, 2012-05-04 for later.

2012-01-10 Simulation program AlloppSim (v3). Update of 2011-10-21 PDF. v3 makes two kinds of BEAST XML file, for AlloppNET (as v2) and AlloppMUL. There are other changes, eg to help with real data.

2012-02-16 Simulation program AlloppSim (v4). Update of 2012-01-10 PDF. v4 adds some analysis code and more code for using real data.

2012-05-04. AlloppNET, AlloppMUL implementation - update. Documentation for my code as it develops. Adding code for arbitrary numbers of diploids and hybridizations to AlloppNET. Includes AlloppMUL implementation.

2012-05-05 Simulation program AlloppSim (v5). Update of 2012-02-16 PDF. v5 extends the analysis code (for counting topologies, log pop values, root heights).

2012-09-01 Simulation program AlloppDT (v1). This describes R code for simulations involving arbitrary numbers of diploids, tetraploids, and hybridizations (`DT'=Diploids+Tetraploids.) It is a modification of AlloppSim (v5).

2012-09-07. AlloppNET implementation - update. This describes the code for a version of AlloppNET which deals with arbitrary numbers of diploids, tetraploids, and hybridizations.

Algorithms

2012-03-30. Algorithm for assigning population values to network nodes. Detailed description of the (complicated) algorithm for assigning population values from an array to network nodes.

2012-05-19 Program for network prior. Description of a program for investigating network priors empirically. Update of 2011-11-23-network-prior-program.pdf.

2012-09-24 Prior, initial state, and MCMC moves for allopolyploid networks. Especially rjMCMC move for changing number of hybridizations. Details of implementation, Hastings ratio calculation.

Simulation results

2012-01-24 Simulation results for two diploids, one hybridization. Six scenarios, two models AlloppNET and AlloppMUL.

2012-05-29 Preliminary results for n diploids, one hybridization. One scenario with 9 diploids, 3 tetraploids.

2012-11-02 Simulation results for a scenario with 13 species Some results for a scenario with 6 diploids, 7 tetraploids, 3 hybridizations, and lots of incomplete lineage sorting.

2012-11-05 Simulation results for 2 diploids, 3 tetraploids Some results for a 3 scenarios with 2 diploids, 3 tetraploids, 1,2,3 hybridizations, and lots of incomplete lineage sorting. The 3 scenarios have similar topologies but different numbers of hybridizations.

Code

2012-01-10. AlloppSim (3rd version). ZIP file containing simulation AlloppSim in R and Seq-Gen source code. See See PDF at 2012-01-10.

2012-02-16. AlloppSim (4th version). ZIP file containing simulation AlloppSim in R, Seq-Gen source code, and support for analysing real data. See See PDF at 2012-02-16.

2012-05-05. AlloppSim (5th version, as used for article). ZIP file containing simulation AlloppSim in R, Seq-Gen source code, and support for analysing real data. Updates from v4 are for producing graphs and tables for article 'Statistical Inference of Allopolyploid Networks in the Presence of Incomplete Lineage Sorting'. See PDF at 2012-05-05.

2012-05-28. R code for real data for article. ZIP file containing R code used to preprocess data for Silene and Pachycladon and generate BEAST XML for AlloppNET and AlloppMUL models. See PDF at 2012-01-14.

2012-02-14 Manual for AlloppNET and AlloppMUL. Manual for using models AlloppNET (2 diploids, one hybridization) and AlloppMUL in BEAST with real data.

Talks

2012-01-10. Trees with legs: phylogenetic analysis of allopolyploids Talk given in Dec 2011 on my work. Slides (PNG images in ZIP file).

2012-01-11. C hoosing models for phylogenetic analysis. Talk given in Dec 2011 during NGS course. Slides (PNG images in ZIP file).