Hi, I've run into your question while looking for something else. This gave me great idea to show my own analysis. I've prepared this fig-5A-like plot and I thought to share it. Here is the result picture:

http://imgur.com/XC9dWXJ

And here is the code (in R):

# Preparing data
data <- data.frame(Gene_set=c("Gene_set1", "Gene_set2", "Gene_set3", "Gene_set4", "Gene_set5"),
                   NES=runif(5, -3, 3),
                   FDR_q.val=runif(5,0,1),
                   No_of_significant_genes=runif(5, 1, 100))

# Plotting
library(ggplot2) 
p <- ggplot(data, aes(NES, Gene_set))
p + geom_point(aes(colour=FDR_q.val, size=No_of_significant_genes)) +
    scale_color_gradientn(colours=rainbow(4), limits=c(0, 1)) +
    geom_vline(xintercept=0, size=0.5, colour="gray50") +
    theme(panel.background=element_rect(fill="gray95", colour="gray95"),
          panel.grid.major=element_line(size=0.25,linetype='solid', colour="gray90"), 
          panel.grid.minor=element_line(size=0.25,linetype='solid', colour="gray90"),
          axis.title.y=element_blank()) +
    expand_limits(x=c(-3,3)) +
    scale_x_continuous(breaks=c(-3,-2,-1,0,1,2,3)) +
    scale_y_discrete(limits=rev(data$Gene_set))

You can get such nice plots using clusterProfiler, as seen here:

http://www.r-bloggers.com/david-functional-analysis-with-clusterprofiler/