Authors: Jonathan Pekar1,2, Michael Worobey3,*, Niema Moshiri4, Konrad Scheffler5, and Joel O. Wertheim6,*
Affiliations:
1Bioinformatics and Systems Biology Graduate Program, University of California San Diego, La Jolla, CA 92093, USA
2Department of Biomedical Informatics, University of California San Diego, La Jolla, CA 92093, USA
3Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ 85721, USA.
4Department Computer Science & Engineering, University of California San Diego, La Jolla, CA 92093, USA
5Illumina, Inc., San Diego, CA 92122, USA
6Department of Medicine, University of California San Diego, La Jolla, CA 92093, USA
*Corresponding authors. Email: worobey@arizona.edu (MW); jwertheim@health.ucsd.edu (JOW)
The main and sensitivity analysis JSON files are used to run the initial FAVITES simulations. Then, for the established epidemics, the CONFIG template file (which invokes VirusTreeSimulator) must be edited to include the proper paths to the contact and transmission networks in the FAVITES outputs from the initial simulations. In the JSON file names, "TF" stands for transmission factor (e.g., "0.33xTF" is 1/3 the transmission rate of the main analysis), "R" is ascertainment rate (e.g., "2xR" is 2-times the ascertainment rate of the main analysis), and "hosp" indicates modulated hospitalization rate (e.g., "hosp2x" is 2-times the hospitalization rate). There can be more than one of these modulations per CONFIG file, as indicated in the name of the file. Please refer to the supplementary tables in the manuscript for further details.
The XML files are the BEAST inputs for the strict, relaxed clock, and Skygrid analyses.
rejection_sampling.py is the code used to generate the main results, and it requires the coalescent data from FAVITES and tMRCAs from BEAST runs.
The main results are in the RejectionSampling folder. All the two-phase results have "TwoPhase" in the name. Particular permutations are mentioned in the name. Please refer to the manuscript for further details.