In datashield/opal: Opal R and Datashield utils

Opal Reporting with R

This is an example of how to use knitr for producing reports on data stored in Opal.

Procedure

The requirement on client side is to have opal package installed:

install.packages('opal', repos=c(getOption('repos'), 'http://cran.obiba.org'), dependencies=TRUE)

The requirements on server side is:

• to have permissions to access to Opal data
• to have permission to send R commands to Opal server
• to have a running R server

The procedure is then the following:

• connect to Opal server
• assign data from Opal to the associated R server
• send to Opal R commands to be executed in R server
• process in R client the data received from these commands

Example

Load the required libraries on client side and login in Opal (credentials and output are hidden)...

library(opal)
o <- opal.login('administrator', 'password', 'http://demo.obiba.org:8080')

Assign some variables into a data.frame with associated D symbol in R on Opal server side:

opal.assign(o,'D','mica_demo.FNAC',variables=c('SVUOSI','SUKUP','PITUUS','PAINO'))

Preview the assigned data.frame:

opal.execute(o,'head(D)')

Summary of the assigned data.frame:

opal.execute(o,'summary(D)')

Histogram figure of the PITUUS variable:

plot(opal.execute(o,'hist(D$PITUUS)'))

Loading ggplot2 library in R on Opal server side... This will fail if ggplot2 is not installed in R server environment.

opal.execute(o, 'library(ggplot2)')

Plot PITUUS vs. PAINO with "lm" smoothing:

opal.execute(o,'qplot(PITUUS,PAINO, data=D) + geom_smooth(method="lm")')

Cleaning the resources on Opal server side...

opal.logout(o)

datashield/opal documentation built on May 14, 2019, 7:53 p.m.