Close

Computation and inference

This theme provides a space for the exploration of ideas for efficient computation, to learn new methodologies for inference and to share knowledge across CMMID.

In the CMMID we use mathematical and statistical tools to understand the dynamics and control of infection. Members use methods of inference to inform data based decisions which can account for large and/or complex data, models and questions. In addition, to deal with these complexities, there is a need for efficient computation. From methods to account for partial observation of cases and uncertainty in confirmation of cases, to tools for creating fast and reproducible code, challenges arise in both computation and inference that are common to many infectious disease research questions.

People

Amanda Minter (theme co-ordinator), Katherine Atkins, Marc Baguelin, Lloyd Chapman, Sam Clifford, Nick Davies, Roz Eggo, Jon Emery, Akira Endo, Flavio Finger, Stefan Flasche, Seb Funk, Liza Hadley, Alasdair Henderson, Chris Jarvis, Petra Klepac, Gwen Knight, Adam Kucharski, Yang Liu, Nicky McCreesh, Hannah Meredith, James Munday, Amy Pinsent, Billy Quilty, Kathleen O’Reilly, Alexis Robert, Tom Sumner, Moritz Wagner, Naomi R Waterlow, Nayantara Wijayanandana, Kevin van Zandvoort

Events

Every Wednesday at 3pm we have an informal walking meeting, the 'Random Walk'. Join us from the corridor outside KS-120 to a nearby cafe to catch up with what your colleagues are working on or ask for help with anything related to computation and inference in your work. 

Methods Roundtable events focus on learning about and discussing the details of a method in infectious disease modelling led by an expert in the field. Previous methods roundtable events have been led by:

  • May 2019Pete Dodd from the University of Sheffield on Gaussian Processes
  • March 2019: Marc Baguelin from LSHTM on Hamiltonian Monte Carlo and Variational Inference using Stan
  • Sep 2018Tom Hladish from the University of Florida on ABC-SMC (Approximate Bayesian Computation – Sequential Monte Carlo)

Within the theme, researchers share their expertise through internal training and introductions to concepts within computation and inference, previous events include:

  • August 2019: Coding workshop. CMMID’s first coding workshop, attendees recreated analysis from a published paper using pairwise coding and collaborative coding techniques. Co-organised with the Early Career Researchers group in CMMID. 
  • July 2019: Introduction to LibBi / RBi. Seb Funk gave an introduction to LibBi and the associated R package RBi used for state-space modelling and Bayesian inference.
  • October 2018: Introduction to data augmentation for infectious disease modelling. Ali Henderson, Amanda Minter, Lloyd Chapman and Kath O’Reilly presented examples of data augmentation from their research (slides available here).

Publications

  • Chatzilena A,, van Leeuwen E.,  Ratmann O., Baguelin M., Demiris, N. (2019). Contemporary statistical inference for infectious disease models using Stan. https://arxiv.org/abs/1903.00423
  • Chapman LAC, Jewell CP, Spencer SEF, Pellis L, Datta S, et al. (2018) The role of case proximity in transmission of visceral leishmaniasis in a highly endemic village in Bangladesh. PLOS Neglected Tropical Diseases 12(10): e0006453. https://doi.org/10.1371/journal.pntd.0006453
  • O’Reilly KM, Cori A, Durry E, Wadood MZ, Bosan A, Aylward RB, et al. (2015) A new method to estimate the coverage of mass vaccination campaigns against poliomyelitis from surveillance data. https://doi.org/10.1093/aje/kwv199 
  • Kucharski AJ, Edmunds WJ (2015) Characterizing the transmission potential of zoonotic infections from minor outbreaks. PLOS Comput Biol 11(4):e1004154
  • Kucharski AJ, Lessler J, Read JM, Zhu H, Jiang CQ et al. (2015) Estimating the life course of influenza A(H3N2) antibody responses from cross-sectional data. PLOS Biol 13(3):e1002082
  • Kucharski AJ, Mills HL, Pinsent A, Fraser C, Van Kerkhove MD et al. (2014) Distinguishing between reservoir exposure and human-to-human transmission for emerging pathogens using case onset data. PLOS Curr. 7:6