covid_and_stat_mech

Creating a situational awareness model for SARS-CoV-2 transmission in Washington state (January 2020 to March 2021).

This is a repository of Python 3.8 code associated with the preprint COVID-19 epidemiology as emergent behavior on a dynamic transmission forest, 2022. In the paper, we create a stochastic process model of SARS-CoV-2 transmission, and we use it to estimate a variety of transmission statistics, from population prevalence to the clustering of cases as outbreaks.

The main scripts are:

1. PathogenesisGPRVis.py, which makes the paper's first and second figures.
2. PrevalenceModel.py, which makes the paper's third figure and a serialized pandas dataframe used as input in subsequent scripts.
3. ContactDistributionsVis.py, which generates the paper's fourth figure.
4. IndividualLevelDistributions.py, which generates the full set of daily contact distributions and serializes some output for use in subsequent scripts.
5. OutbreakInvestigation.py, which makes Figure 5a from the paper.
6. TransmissionForestVis.py, which makes Figure 5b from the paper.
7. ForestStats.py, which makes Figure 5c from the paper.

November 2023 update

We've added to this repository additional code associated with our follow-up preprint A generating function perspective on the transmission forest, 2023. In that paper, we use the Washington state model above as an example to illustrate the application of a formal, generating function-based approach to transmission forest calculations.

The main scripts for this second paper are

1. generating_functions\SampleTransmissionTrees.py, which creates a large serialized set of sample trees for method validation.
2. generating_functions\SamplingVis.py, which makes the paper's first figure.
3. generating_functions\ConstellationDists.py, which makes the paper's second figure.
4. generating_functions\FormalimVsSamples.py, which makes the paper's third figure.

building the Python environment

The Python environment is managed through conda and described in environment.yml. Within an Anaconda Prompt, type conda env create python=3.8 -f environment.yml. This will create an environment named covidx.

If your conda installation is not at the default path or you'd prefer to create it elsewhere, change the prefix path in environment.yml before running the command above.

If you run into errors with the Qt platform plugin when trying to run scripts from the covidx environment, such as,

qt.qpa.plugin: Could not find the Qt platform plugin "windows" in ""
This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.

here's the solution that worked for me, based on this issue reply:

Copy the content of [Anaconda directory]\envs\covidx\Library\plugins\platforms to [Anaconda directory]\envs\covidx\platforms.