Single_strain_model

This directory contains code to estimate the model parameters for a given subtype (either H3N2 or H1N1pdm09), to simulate the latent dynamics of infection, and to perform the model validation. The work-flow follows below. Please refer to the Introductory page for a step-by-step overview of the manuscript analysis.

1. Perform the model inference (using the code in the Inference subdirectory): a. Execute the make_host_data_list.R script to generate a list of host data objects b. Convert the host data list from step 1 to a POMP object using the data_to_pomp_object_structure.R file. c. Execute global searches of the likelihood space and parameter profiles (see the detailed explanation on the Introductory page) to estimate the model parameters.

2. Analyze the inference results and calculate the parameter MLEs/95% CIs (using the code in the Results_analysis subdirectory). After you have estimated all of the fitted parameters, create a .rda file containing a named vector of all parameters for the model simulations (step 3).

3. Simulate the dynamics from the best-fit model parameters, generate the output plots, and perform model validation (using code in the Simulations subdirectory).