Add ylabel and point shape change in ggpairs

Question

ggpairs provide graph with two box plots connecting points to show before and after situation after giving an intervention.

Want to change y axis label "Resist" "Suscept" instead of "3GCr" and "3GCs".
And want to change data point with r default shape like Resist= circle (1) & Suscept= Diamond(5)

Searched several sites but not get the desired result. Also add code and output:

df.reach.sub <- data.frame(Resist = c(5.78, 3.85, 5.60, 7.83, 6.68, 7.90), Suscept =c(5.70, 3.70, 8.82, 8.67, 9.06, 8.08), Age = c(1,1,1,1,1,1), Status = c("Positive", "Positive", "Positive", "Positive", "Positive", "Positive"))

df.reach.subCat1 <- df.reach.sub[which(df.reach.sub$Age == "1-3" & df.reach.sub$Status == "Positive"),]
gp.plotCat1 <- ggpaired(df.reach.subCat1, cond1 = "Resist", cond2 = "Suscept", color = "Black", line.color = "gray", line.size = 0.4, palette = "npg", ylab = expression('log'[10]*' CFU / g faeces'), xlab = NULL, main = expression("A) 1-3 months ("*italic('p = 0.116, n = 6')*')'), legend = "none", ggtheme = theme_bw(), ylim = c(2,10))

gp.plotCat1

Welcome to SO. Please provide an MCVE ([here](https://stackoverflow.com/help/mcve)), this will help us to help you. — tjebo, Aug 09 '18 at 06:46
have a look at this thread : https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example. the best is to take a bit of time and create a new , minimal dataset (or use one of the inbuilt ones) — tjebo, Aug 09 '18 at 07:58
your question 1) is unclear to me. Your y-axis is not labelled as you have suggested. Question 2) A look into the ggpairs documentation suggests that you might be able to do this changing the `mapping` argument. On the other hand, your plot could easily be done with `ggplot2` only. Maybe this could be an option for you — tjebo, Aug 12 '18 at 11:45
@Tjebo I know ggplot2 is another good option but want to do this ggpaired. Also checked ggpaires doc but could not find anything to change x axis and value label shape change. — Md. Rayhanul Islam, Aug 13 '18 at 08:15

score 0 · Answer 1 · answered Nov 05 '18 at 09:23

I recommend to use some tidyverse and "basic" ggplot2, like this

df.reach.sub %>% 
  rownames_to_column() %>% 
  gather(key, value, -Age, -Status, -rowname) %>% 
  ggplot(aes(key, value)) +
    geom_boxplot() + 
    geom_point(aes(shape=key), show.legend = F, size=2) + 
    geom_path(aes(group=rowname)) + 
    theme_bw()

Add ylabel and point shape change in ggpairs

1 Answers1