Modelling the spread of COVID-19 in England
June Dalziel Almeida
JUNE
Arnau Quera-Bofarull
Durham University CDT PhD student
Who are we?
Arnau Quera-Bofarull
Aidan Sedgewick
Miguel De Icaza Lizaola
Carolina Cuesta-Lazaro
Henry Truong
Astronomy / Particle Physics CDT students
Amazon
Joseph Aylett-Bullock
United Nations
Procter & Gamble
HEP Inspire
Boeing
Ibex
Kevin Fong
the (actual) doctor
Ian Vernon
the statistician
Tristan Caulfield
the hacker
Bryan Lawrence
the climate scientist
Julian Williams
the business man
Frank Krauss
the particle physicist
Who are we?
Other Researchers
What is JUNE?
England Digital Twin
Epidemiological model
Agent-based simulation
England Digital twin
Geography
England Digital twin
Demography
- age (27)
- sex (f)
- ethnic group (Caribbean)
- deprivation index 2 (1-10)
- work sector / subsector (healthcare/doctor)
- mode of transport (public)
- area of residence
- super area of work
Main data source: census data (NOMIS)
male
female
granular Demography
male
female
England Digital twin
dynamics (where can people go?)
- Residence
- Care Home
- Household
- Primary activity
- Company
- Hospital
- School
- Care Home
- University
-
Travel
- Commute
- National travel
- Leisure
- Shopping
- Pubs / restaurants
- Cinema
- Household visits
City of London workers' usual residence
Matching workplace and residence
places are geolocalized
DURHAM STUDENT POPULATION
43 yo
38 yo
10 yo
How do people transmit the infection?
j
j
i
Intensity of contacts (per group)
Infectiousness profile
Contact
Matrix
Disease Trajectory
Odds calibrated to data
POlicies
MODEl fitting
We have
Model (18 free parameters)
+
Data (quite messy)
It takes ~400 CPU hours to run one simulation....
How do we fit it?
bayesian emulator + history matching
bayesian emulator + history matching
Train Bayesian emulator
Run emulator
O(500k) times
Run full simulation O(100) times
Narrow parameter space search
IT WORKS!
checkout the code at
github.com/IDAS-Durham/JUNE
