This package is under development. Do not use it without contacting the authors first.
To install the current stable, CRAN version of the package, type:
install.packages("simulacr")To benefit from the latest features and bug fixes, install the development, github version of the package using:
devtools::install_github("reconhub/simulacr")Note that this requires the package devtools installed.
simulacr implements outbreak simulation using branching processes. The main features of the package include:
-
simulate_outbreak: the main function, allowing to simulate outbreaks for specified values of reproduction number, incubation period, duration of infectiousness, and optionally reporting delays; it outputs a linelist stored as adata.framewith the classoutbreak, including information on transmission chains; the output can be converted toepicontactsobjects for visualisation, and plotted usingplot(...) -
as_epicontacts: a function to convertoutbreakobject toepicontact; called implicitely when plottingoutbreakobjects -
make_disc_gamma: wrapper function to build discretised Gamma distributions easily -
draw_labels: wrapper function to generate random alphanumeric labels
This brief example illustrates how we can simulate an outbreak, with different ways of specifying input distributions.
Distributions can be specified in different ways:
- as a discretised distribution stored as
distcreteobject, e.g. generated viamake_disc_gamma - as a function with a single argument (the delay) computing probability mass function (pmf) of different delays
- as a vector of
$n$ positive numbers, taken to be the pmf of delays of 0:($n$ -1)
We illustrate these different options below, with:
- a flat distribution of the incubation time 1-4 days (all other values have a null probability)
- an infectious period defined as a discretised Gamma distribution stored as a
distcreteobject - a reporting delay following a Poisson distribution, specified as a function
library("simulacr")
incubation <- c(0, 1, 1, 1, 1) # numbers = unscaled PMF
infectious_period <- make_disc_gamma(10, 7) # distcrete object
reporting <- function(x) dpois(x, 5) # PMF functionset.seed(1)
x <- simulate_outbreak(R = runif(100, 1, 3), # random values on [1;3]
dist_incubation= incubation,
dist_infectious_period = infectious_period,
dist_reporting = reporting)The output is a data.frame with the class outbreak, which contains a
linelist of cases:
class(x)## [1] "outbreak" "data.frame"
dim(x)## [1] 77 6
head(x)## id source date_infection date_onset date_report R
## 1 eLGzdD <NA> 0 3 12 1.954460
## 2 MQ90i6 eLGzdD 4 5 12 2.209867
## 3 VCEaZ4 MQ90i6 8 11 18 2.284577
## 4 bEGjSA eLGzdD 12 13 17 2.321596
## 5 k6rcYK eLGzdD 12 16 19 1.764776
## 6 msP3Ki bEGjSA 15 16 19 2.514174
tail(x)## id source date_infection date_onset date_report R
## 72 K0450K S0jiBe 83 84 89 2.321596
## 73 7a20Ap xeKC7z 83 86 91 1.403364
## 74 41R9ih uijTrA 85 87 91 1.820168
## 75 oiHVQb Q2diLw 86 87 91 2.284577
## 76 vEGjkv 7a20Ap 86 88 92 2.796779
## 77 uo5Bds shTo1z 90 94 98 1.490977
This object can be plotted using (this will open an interactive graph:
plot(x)
For any work relying on transmission trees, it may be easiest to convert the
outbreak object to an epicontacts:
net <- as_epicontacts(x)
net##
## /// Epidemiological Contacts //
##
## // class: epicontacts
## // 77 cases in linelist; 77 contacts; directed
##
## // linelist
##
## # A tibble: 77 x 4
## id date_infection date_onset date_report
## <chr> <dbl> <int> <int>
## 1 eLGzdD 0 3 12
## 2 MQ90i6 4 5 12
## 3 VCEaZ4 8 11 18
## 4 bEGjSA 12 13 17
## 5 k6rcYK 12 16 19
## 6 msP3Ki 15 16 19
## 7 Ntdjbt 16 17 21
## 8 iwXiO6 18 20 24
## 9 vHa6qJ 18 19 22
## 10 8Fix6l 19 21 23
## # … with 67 more rows
##
## // contacts
##
## # A tibble: 77 x 2
## from to
## <chr> <chr>
## 1 <NA> eLGzdD
## 2 eLGzdD MQ90i6
## 3 MQ90i6 VCEaZ4
## 4 eLGzdD bEGjSA
## 5 eLGzdD k6rcYK
## 6 bEGjSA msP3Ki
## 7 bEGjSA Ntdjbt
## 8 bEGjSA iwXiO6
## 9 MQ90i6 vHa6qJ
## 10 VCEaZ4 8Fix6l
## # … with 67 more rows
Contacts can be simulated and added to a simulated outbreak, using similar
procedures to the one used in simulate_outbreak, with a few differences:
-
n_contacts: the numbers of contacts per index case is specified in the same way as the reproduction numberRinsimulate_outbreaks -
the two distributions relate to the time between onset of the index case and the beginning of secondar exposures (
dist_time_to_contact) and the duration of the exposure window (duration) after the first day of contact
## exposure starts 0-2 days post onset
time_to_contact = c(1, 1, 1)
## geom dist for duration of exposure
duration <- function(x) dgeom(x, prob = .9)
x_with_contacts <- simulate_contacts(
x[1:10, ],
n_contacts = 1:10, # 1 to 10 contacts
dist_time_to_contact = time_to_contact,
dist_duration = duration)
## check output
class(x_with_contacts)## [1] "outbreak" "data.frame"
dim(x_with_contacts)## [1] 64 9
head(x_with_contacts)## id source date_infection date_onset date_report R type
## 1 eLGzdD <NA> 0 3 12 1.954460 case
## 2 MQ90i6 eLGzdD 4 5 12 2.209867 case
## 12 x72VZq eLGzdD NA NA NA NA non_case
## 14 tsI8G0 eLGzdD NA NA NA NA non_case
## 11 HBGWgN eLGzdD NA NA NA NA non_case
## 13 ewV5zO eLGzdD NA NA NA NA non_case
## date_exposure_start date_exposure_end
## 1 0 0
## 2 4 4
## 12 4 4
## 14 4 4
## 11 5 5
## 13 5 5
plot(x_with_contacts)
No vignette currently available.
Bug reports and feature requests should be posted on github using the issue system. All other questions should be posted on the RECON forum:
http://www.repidemicsconsortium.org/forum/
Contributions are welcome via pull requests.
Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.