Make tooltip appear on corresponding points in different ggplot2 facets in a plotly object

Question

I am using ggplot2 and the plotly R packages to generate a volcano plot to visualize protein differential abundance data.

library(ggplot2)
library(plotly)

nVals <- 80
nFacets <- 2

#example dataset
proteins <- rep(paste0('protein_', c(1:(nVals / nFacets))), nFacets)
set.seed(1)
dat <- data.frame(log_FC = c(rnorm(nVals*0.8, 0, 1), rnorm(nVals*0.2, 0, 12)),
                  log_Pval = abs(rnorm(nVals, mean=0, sd=0.01)),
                  facet = rep(paste0('Cell line ', 1:nFacets), nVals / nFacets),
                  protein = proteins[order(proteins)])


#make ggplot2 object
p <- ggplot(dat, aes(y = log_Pval, x = log_FC, text = protein)) +
  facet_wrap(~ facet) +
  geom_point()

#convert p to plotly object with plotly::ggplotly
ggplotly(p)

Example of current output

I am using the tooltip feature in plotly to show the data associated with each point. I would like to extend the tooltip feature to highlight the same protein in different facets of the plot.

In other words, when a cursor hovers over a point in 1 facet, the tooltip box would show up on all points which have the same value in the dat$protein column in adjacent facets.

Here is an example of what I am aiming for.

Is there some way of the customizing the behavior of the tooltip to achieve the what I have described?

Answer 1

With crosstalk, you can have widgets communicate with each other. Create a SharedData object from your data frame, and select protein as the key.

library(crosstalk)
shared_df <- SharedData$new(dat, key = ~protein)

Then, use shared_df instead of dat with ggplot. If you select a point in one plot, it will highlight the matching data point (by protein) in the second plot. Please let me know if this meets your needs.