Is there a CHIME mode for COVID-19 by R-Studio?

[Visiting]

There is a Mode for "COVID-19 Hospital Impact Model for Epidemics (CHIME)", here is the link for it:
https://penn-chime.phl.io/

The code for this model can be found here:

chime/src/penn_chime at develop · CodeForPhilly/chime

But, it is python code and has to be run at linux environment, is there a similar model that can be run by R-Studio? Thank you!

[technocrat]

RStudio can run Python!

If you decide to go that way, please post a new question on reticulate related matters if you run into difficulties. I've been away from Python for longer than I realized before thinking about it just now.

[Visiting]

Thank you! I will try.

[Visiting]

I created a R code below and try to create similar curve for "Susceptible, Infected, and Recovered", like the third graph here: https://penn-chime.phl.io/
But, I don't know how to determine the parameters beta, gamma and others. Can anyone help to generate those parameter based on the information provided from the website? Thank you.

install.packages(deSolve)
library(deSolve)

writing the differential equations in R

sir_equations<- function(time, variables, parameters) {
with(as.list(c(variables, parameters)), {
dS<- -beta * I * S
dI<- beta * I * S - gamma * I
dR<- gamma * I
return(list(c(dS, dI, dR)))
})
}

defining some value for the parameters

parameters_values<- c(
beta = 0.001, # infectious contact rate (/person/day)
gamma = 0.015 # recovery rate (/day)
)

initial values for the variables

initial_values<- c(
S=3599733, # number of susceptibles at time = 0
I=266, # number of infectious at time =0
R=0 # number of recovered (and immune) at time=0
)

time_values<- seq(0, 125) # days

ode() function

sir_values_1 <- ode(
y = initial_values,
times = time_values,
func = sir_equations,
parms = parameters_values
)

sir_values_1 <- as.data.frame(sir_values_1)

plot

with(sir_values_1, {
plot(time, S, type = "l", col = "blue", xlab = "time (days)", ylab = "number of people")
lines(time, I, col = "red")
lines(time, R, col = "darkgreen")
})

legend("right", c("susceptibles", "infectious", "recovered"),
col = c("blue", "red", "darkgreen"), pch = 1, bty = "n")

[technocrat]

Hi, Jerry,

Please see the FAQ: What's a reproducible example (`reprex`) and how do I do one? Using a reprex, complete with representative data will attract quicker and more answers. Also, in the reprex it's good etiquette to use

require(deSolve) # in place of install.packages("deSolve")

The parameters are user selected arguments to the sir_values_1 function, which self destructs with

sir_values_1 <- as.data.frame(sir_values_1)

You probably want to do something like

run1 <-  as.data.frame(sir_values_1)

So, to recover those

# require(deSolve) # in place of install.packages(deSolve)

### snip


parameters_values<- c(
beta = 0.001, # infectious contact rate (/person/day)
gamma = 0.015 # recovery rate (/day)
)

### snip

### added to show where to find beta and gamma parameters; none others used except for starting values 

parameters_values
#>  beta gamma 
#> 0.001 0.015

^{Created on 2020-04-05 by the reprex package (v0.3.0)}

The choice of which parameter values is domain specific. All assumptions within the permitted range of a function to be given as arguments are equally valid from a data science computational stance; whether they are equally reasonable is a matter of the analyst's judgment about the subject matter.

This community may have some epidemiologists from time-to-time, but I imagine that most of them are, sadly, too busy now to offer advice. I suggest moving the types of questions beyond the mechanics and going into the realm of subject matter expertise into more general online discussions.

Good luck! And, thanks for sharing Chimes! Extremely interesting.

P.S. check out this page on the CHIME site

[Visiting]

Thank you so much for your suggestion and information!