Skip to content

Latest commit

 

History

History
54 lines (35 loc) · 2.85 KB

readme.md

File metadata and controls

54 lines (35 loc) · 2.85 KB
Raw

Data and scripts from the study "Assessing the Adequacy of Morphological Models used in Palaeobiology"

1. Empirical-Inf

This contains everything for the empirical analysis. Data directory contains all the morphological data sets analysed. Scripts directory contains all revbayes scripts for inference and Rscripts for down stream analysis.

Start.sh will run all data sets under the 7 models.

2. PPS Simulations

This contains the set up for the simulation study.

Simulation

The data directory contains the output from the empirical inference of Egi and Shoshani for four different models. Simulated data sets are in the data_model directories. Scripts contains all the revbayes scripts used for the inference. Data was simulated using the Sim.r file. To simulate data, you need to specify the model, data set, and runID. This requires editing the Sim.r file directly, lines 41-45

Analysis

Contains two directories, Egi and Shsoahni. Each contains the posterior predictive simulations’ workflow. To start the analysis run the scripts in the jobs_"model" files. All revbayes scripts used are in the scripts folder. There are five other scripts used in the analysis

  • sim-start.sh: this contains all the commands for each individual run
  • CheckConvergence.r: this file ensures that the initial MCMC has reach convergence before simulating new data sets
  • MorphoSim.r: simulates data sets in R using the phangorn R package.
  • Anaylsis.r: this file calculates the tests statistics. Need to specify on line 3 and 11 which simulation set up (which model) you are analysing.
  • Cumulative.r: this file was used to assess the number of replicates data sets are necessary to determine the adequacy of a model. This was only used once, after which we simulated 500 data sets for the rest of the models
Stepping Stone

Contains two directories, Egi and Shsoahni. Each contains the scripts for the stepping stone analysis. To start the analysis run the scripts in the jobs_"model" files. All revbayes scripts used are in the scripts folder. There are two other scripts used in the analysis

  • stepping_stone.sh: his contains all the commands for each individual run
  • ml_ss.sh: gets all the marginal likelihoods and puts them into a cvs file. Requires you provide the model name as an argument.

3. PPS Empirical

The scripts required to carry out posterior predictive simulations on the empirical data sets. All the revbayes scripts are in the scripts directory. Running the files in the jobs_"datasetname" directories will start the analysis. There are four other scripts used in the analysis

  • MorphoSim.r: simulates data sets in R using the phangorn R package.
  • Test_Statistics.r: calculates all test statistics
  • Fig-6.r: calculates consistency index and retention index test stats
  • P-vals.r: calculates the posterior predictive p-values