Adding a table to ggplot figure

Question

I have dose response data:

df <- data.frame(dose=c(10,0.625,2.5,0.15625,0.0390625,0.0024414,0.00976562,0.00061034,10,0.625,2.5,0.15625,0.0390625,0.0024414,0.00976562,0.00061034,10,0.625,2.5,0.15625,0.0390625,0.0024414,0.00976562,0.00061034),viability=c(6.117463479317,105.176885855348,57.9126197628863,81.9068445005286,86.484379347143,98.3093580807309,96.4351897372596,81.831197750164,27.3331232120347,85.2221817678203,80.7904933803092,91.9801454635583,82.4963735273569,110.440066995265,90.1705406346481,76.6265869905362,11.8651732228561,88.9673125759484,35.4484427232156,78.9756635057238,95.836828982968,117.339025930735,82.0786828300557,95.0717213053837),stringsAsFactors=F)

I fit log-logistic model to these data using the drc R package:

library(drc)
fit <- drm(viability~dose,data=df,fct=LL.4(names=c("Slope","Lower Limit","Upper Limit","ED50")))

I then plot this curve with the standard error using:

pred.df <- expand.grid(dose=exp(seq(log(max(df$dose)),log(min(df$dose)),length=100))) 

pred <- predict(fit,newdata=pred.df,interval="confidence") 
pred.df$viability <- pred[,1]
pred.df$viability.low <- pred[,2]
pred.df$viability.high <- pred[,3]



library(ggplot2)
p <- ggplot(df,aes(x=dose,y=viability))+geom_point()+geom_ribbon(data=pred.df,aes(x=dose,y=viability,ymin=viability.low,ymax=viability.high),alpha=0.2)+labs(y="viability")+
geom_line(data=pred.df,aes(x=dose,y=viability))+coord_trans(x="log")+theme_bw()+scale_x_continuous(name="dose",breaks=sort(unique(df$dose)),labels=format(signif(sort(unique(df$dose)),3),scientific=T))+ggtitle(label="all doses")

Finally I'd like to add the parameter estimates to the plot as a table. I'm trying:

params.df <- cbind(data.frame(param=gsub(":\\(Intercept\\)","",rownames(summary(fit)$coefficient)),stringsAsFactors=F),data.frame(summary(fit)$coefficient))
      rownames(params.df) <- NULL

ann.df <- data.frame(param=gsub(" Limit","",params.df$param),value=signif(params.df[,2],3),stringsAsFactors=F)
rownames(ann.df) <- NULL
xmin <- sort(unique(df$dose))[1]
xmax <- sort(unique(df$dose))[3]
ymin <- df$viability[which(df$dose==xmin)][1]
ymax <- max(pred.df$viability.high)
p <- p+annotation_custom(tableGrob(ann.df),xmin=xmin,xmax=xmax,ymin=ymin,ymax=ymax)

But getting the error:

Error: annotation_custom only works with Cartesian coordinates

Any idea?

Also, once plotted, is there a way to suppress row names?

Community · Accepted Answer · 2017-05-23 11:48:43Z

7

I'm not sure if there's a way around the annotation_custom error within "base" ggplot2. However, you can use draw_grob from the cowplot package to add the table grob (as described here).

Note that in draw_grob the x-y coordinates give the location of the lower left corner of the table grob (where the coordinates of the width and height of the "canvas" go from 0 to 1):

library(gridExtra)
library(cowplot)

ggdraw(p) + draw_grob(tableGrob(ann.df, rows=NULL), x=0.1, y=0.1, width=0.3, height=0.4)

Another option is to resort to grid functions. We create a viewport within the plot p draw the table grob within that viewport.

library(gridExtra)
library(grid)

Draw the plot p that you've already created:

Create a viewport within plot p and draw the table grob. In this case, the x-y coordinates give the location of the center of the viewport and therefore the center of the table grob:

vp = viewport(x=0.3, y=0.3, width=0.3, height=0.4)
pushViewport(vp)
grid.draw(tableGrob(ann.df, rows=NULL))

UPDATE: To remove the background colors of the table grob, you can manipulate the table grob theme elements. See the example below. I also justified the numbers so that they line up on the decimal point. For more information on editing table grobs, see the tableGrob vignette.

thm <- ttheme_minimal(
  core=list(fg_params = list(hjust=rep(c(0, 1), each=4), 
                             x=rep(c(0.15, 0.85), each=4)),
            bg_params = list(fill = NA)),
  colhead=list(bg_params=list(fill = NA)))

ggdraw(p) + draw_grob(tableGrob(ann.df, rows=NULL, theme=thm), 
                      x=0.1, y=0.1, width=0.3, height=0.4)

edited May 23, 2017 at 11:48

CommunityBot

11 silver badge

answered Oct 31, 2016 at 2:45

eipi10

94.6k28 gold badges220 silver badges299 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

dan Over a year ago

eipi10, is it possible to control the transparency of the table added using draw_grob function? in some cases I wont have a clear white background to draw the table over and would it to be transparent so that what's behind it in the figure is not completely masked.

Collectives™ on Stack Overflow

Adding a table to ggplot figure

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related