In reconhub/vimes: VIsualisation and Monitoring of EpidemicS

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>", 
  fig.width = 6, 
  fig.height = 6, 
  fig.path = "figs-demo/"
)

This vignette provides a short demonstration of the package using a dummy dataset.

We first simulate the data using 3 mixtures of 3 normal distributions, and compute Euclidean distances between the observations for each mixture. In practice, each mixture would be a different data type (e.g. location, time of onset of symptoms, genetic sequences of the pathogen):

set.seed(2)
dat1 <- rnorm(30, c(0,1,6))
dat2 <- rnorm(30, c(0,0,1))
dat3 <- rnorm(30, c(8,1,2))
x <- lapply(list(dat1, dat2, dat3), dist)

The function vimes_data processes the data and ensures matching of the individuals across different data sources:

library(vimes)

x <- vimes_data(x)
plot(x)

We can now run vimes on the data:

res <- vimes(x, cutoff = c(2,4,2))
names(res)

res$graph
res$clusters

The main graph is:

plot(res$graph, main="Main graph")
for(i in 1:3) {
plot(res$separate_graphs[[i]]$graph, main = paste("Graph from data", i))
}

reconhub/vimes documentation built on May 27, 2019, 4:03 a.m.