The Prisoner of War (PoW) model is written in R, allows for the reconstruction of virus time trees over deep timescales, and is based on a mechanistic evolutionary model that explains the time-dependent changes in the evolutionary rate estimates of viruses.
PoW model requires the posterior rate distribution from dated sequences in the dataset. It also requires an ultrametric distance tree of all sequences in the dataset using a standard HKY substitution model (with no assumed rate heterogenity) along with the posterior distributions of the base frequencies and transition-transverstion rate. All output files should be compatible with the BEAST 1 format.
-
*.logincludes the posterior rate distribution (meanRate column in the log file) of the dates samples -
*.logincludes the posterior base frequencies and transition-transversion rate distributions (frequencies1-4 and kappa columns in the log file) of the entire dataset using a standard HKY substitution model -
*.treesincludes the posterior ultrametric distance trees
PoW model is compatible with R version 4.0.0 or higher and is tested on 4.0.4 to 4.2.1. It depends on several libraries:
-
ape
-
nleqslv
-
stringr
Ensure all required libraries are installed and the prerequisited files are available on your local directory.
Ensure that in the PoW_mode.R, set all the paths to the corresponding directories where the prerequisited files are located -- these can be found in the script lines #9, #13, #14, #95, #98, and #199.
- Copyright and License: Mahan Ghafari, MIT Licence
- Reference:
- A mechanistic evolutionary model explains the time-dependent pattern of substitution rates in viruses by Mahan Ghafari, Peter Simmonds, Oliver G Pybus, and Aris Katzourakis. Current Biology (2021).