Python基于SEIR传染病模型和MCMC马尔可夫链蒙特卡洛算法的疫苗接种场景模拟仿真源码

# Modelling COVID-19 dynamics and potential for herd immunity by vaccination in Austria, Luxembourg and Sweden ## Description: This code represents the implementation of our extended SEIR model with ICU, hospital, recovering, death and vaccination. With this code, we generated the results and simulations of the manuscript [1] published on Journal of Theoretical Biology and available at https://doi.org/10.1016/j.jtbi.2021.110874. ## Structure of the code (Pseudocode): 1) Importing dataset 2) Computing seven days moving average 3) Creation of a calender for the plots 4) If needed: part for fitting procedure of the prevelance data in order to find a function p1 5) Just for Sweden: fitting procedure in order to find a p2 function to describe the behavior 6) ODE system based on SEIR 7) Introduction of social interaction function as a step-wise function 8) Solving of ODE system 9) Plotting and simulation of three different social interaction-scenarios (cumulative cases, daily cases, death, ICU and hospital) 10) Likelihood function for the different waves 11) Parallel MCMC 12) Gelman-Rubin diagnostic 13) 2D density plot 14) Fold change plot 15) Vaccination part with different vaccination strategies 16) Simulation of different vaccination scenarios 17) Heat map for herd immunity by April 18) Heat map for herd immunity over time ## Table of Contents: This repository contains different types of formats. We have different excel sheets which contain the data until 15th December. - The excel file: Sweden_data.xlsx contains the data for Sweden until 15th December. - The excel file : Austria.xlsx contains the data for Austria until 15th December. - The excel file: Lux-trial_15December.xlsx contains the data for Luxembourg until 15th December. - The Jupyter notebook: Luxembourg_Residents_Fit_1st_and_2nd_wave_Optimize_and_MCMC_ParallelFullChains.ipynb is the code for Luxembourg. - The Jupyter notebook: new_model_for_Sweden_minimize-latest-version.ipynb is the code for Sweden. - The Jupyter notebook: Austria_main_model.ipynb is the code for Austria. ## How to run the scripts To run our code: - Download the full content of the directory containing this README file. - Make sure you have installed Jupyter (we developed the code under version: jupyter-notebook : 6.1.4, so we suggest this or subsequent versions). Additionally, the code has been tested on macOS Sierra 10.12.6 and should run on every OS. - Open Jupyter. - Select the script corresponding to the analysis you are interested in running, and run it. We used the following libraries and version of the libraries : - sys==Python BuiltIn - xlrd==1.2.0 - seaborn==0.9.0 - scipy==1.5.2 - pymcmcstat==1.9.0 - pymc3==3.8 - pandas==1.1.1 - numpy==1.19.1 - mcmcplot==1.0.1 - matplotlib==3.3.1 - lmfit==1.0.1 More information of what concretly is used in relation to Jupyter: - jupyter core : 4.6.3 - jupyter-notebook : 6.1.4 - qtconsole : 4.7.6 - ipython : 5.1.0 - ipykernel : 5.3.4 - jupyter client : 6.1.6 - jupyter lab : 2.2.8 - nbconvert : 5.4.1 - ipywidgets : 7.5.1 - nbformat : 5.0.7 - traitlets : 4.3.3 ## Credits To use our original or adapted codes, please cite our work https://doi.org/10.1016/j.jtbi.2021.110874 as [1], see reference at the end of this document. The Jupyter notebook code present in this directory has been written by Françoise Kemp and Stefano Magni, the overall project received contributions from Daniele Proverbio, Laurent Mombaerts, Atte Aalto, Aymeric Fouquier d'Hérouël, Andreas Husch, Christophe Ley, Jorge Gonçalves and Alexander Skupin. This work can be found as a paper in the manuscript published in 2021 on Journal of Theoretical Biology https://doi.org/10.1016/j.jtbi.2021.110874 and titled "Modelling COVID-19 dynamics and potential for herd immunity by vaccination in Austria, Luxembourg and Sweden". The first version of this work appeared on MedrXiv in January 2021 as preprint: https://doi.org/10.1101/2020.12.31.20249088. [1] Françoise Kemp, Daniele Proverbio, Atte Aalto, Laurent Mombaerts, Aymeric Fouquier d’Hérouël, Andreas Husch, Christophe Ley, Jorge Gonçalves, Alexander Skupin, Stefano Magni, "Modelling COVID-19 dynamics and potential for herd immunity by vaccination in Austria, Luxembourg and Sweden", Journal of Theoretical Biology, Volume 530, 2021, 110874, ISSN 0022-5193, https://doi.org/10.1016/j.jtbi.2021.110874