Skip to content

kingaa/sbied

Repository files navigation

Simulation-based Inference for Epidemiological Dynamics
Course Materials

These are notes from a short course given by Ed Ionides and Aaron King at the Summer Institute in Statistics and Modeling in Infectious Diseases (SISMID). From the main page, links lead to pages on a number of specific topics, culminating in four case studies that exemplify the methods and raise key issues. For each such page, there is a corresponding R script, which contains the codes needed to recapitulate the calculations. This script is a starting point for students to follow, explore, and modify the analysis according to their own curiosity and interest.


Required software

The codes require, at a minimum, R version 4.2 and pomp version 5.2. Windows users must also have the appropriate version of Rtools installed. The prep directory contains scripts that will install other needed packages and test the user’s installation.

The following R packages are needed to build the materials, but are not used in the materials:

  • revealjs
  • diagram
  • mapproj

The contacts lesson requires panelPomp. An installation script is provided there.

A Github Action checks that these installations and actions succeed on a variety of current and legacy platforms:

install-test


Manifest

  • index.md, index.html: Landing page
  • acknowledge.md, acknowledge.html: Acknowledgements
  • subdirectories:
    • prep: Lesson 0, on preparing for the course
    • intro: Lesson 1
    • stochsim: Lesson 2
    • pfilter: Lesson 3
    • mif: Lesson 4
    • measles: Lesson 5
    • polio: Lesson 6
    • ebola: Lesson 7
    • contacts: Lesson 8
    • od: Lesson from an earlier version of the course
    • misc: Miscellaneous lessons
    • _includes: latex, HTML, R files included in other files
    • graphics: figures in various formats
  • LICENSE.md: Full text of license
  • TODO.md: To-do list
  • Makefile, rules.mk: GNU Make files
  • sbied.bib: Bibliography database
  • jss.bst, jss.csl: Bibliography style files
  • CLUSTER.R: example CLUSTER.R file for site-specific cluster setup

Compilation

Full compilation of all the materials can be accomplished by running make in the root directory. This requires substantial resources. For this reason, the most expensive computations are archived using the facilities provided for the purpose in pomp. Compilation with these archives in place requires much less time than does compilation from scratch. The following gives an indication of the size of the archives and the time required for their computation on a linux cluster with 250 cpus.

directory size (kB) time (min)
measles/results 340 224.0
mif/results 2244 67.6
polio/results 41687 65.6
polio/initial-values-exercise/results 41635 64.3
polio/starting-values-exercise/results 41687 64.2
contacts/results 206 13.7
pfilter/results 142 13.2
polio/runlevel2/results 3266 4.8
polio/runlevel1/results 612 0.8
ebola/results 2259 0.4

The archives amount to about 131 MB. About 126 MB of this is associated with the polio lesson, in which large amounts of redundant information are stored.

Full compilation, i.e., rebuilding the complete set of materials following deletion of all archives, requires about 534 min on a 250-processor cluster. Full compilation regenerates the complete set of archives. A finishing compilation, i.e., rebuilding with all archives in place, but with the re-Making of all documents, requires about 12 min on a 64-cpu workstation.


License

CC-BY-NC

This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). Under its terms, you are free to:

  • Share — copy and redistribute the material in any medium or format
  • Adapt — remix, transform, and build upon the material

under the following terms:

  • Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
  • NonCommercial — You may not use the material for commercial purposes.
  • No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.

The licensor cannot revoke these freedoms as long as you follow the license terms.


About

Simulation-based Inference for Epidemiological Dynamics

Resources

License

Stars

Watchers

Forks

Languages