Generating horizontal notch plot in ggplot2

Question

I have a data with 3 different disease conditions and several outcomes of them. Diseases are in the first column and outcomes are named from second column on. Please see the data in the attachment.

Disease A   B   C   D   E   F   G   H
X   0.001527882 0.003459413 0.000648574 0.002127627 0.000847073 0.002628663 0.00000156  0.001697012
X   0.001848781 0.003641997 0.000876108 0.0024079   0.001061678 0.00253612  0.00000191  0.001761288
X   0.004304498 0.005887634 0.003197471 0.003581047 0.002125073 0.003125686 0.000025    0.001974246
X   0.006675181 0.00784345  0.004085157 0.005653283 0.003118116 0.001897106 0.000011    0.004055166
X   0.006206229 0.007407223 0.003775627 0.005398474 0.002903612 0.001901196 0.0000138   0.003867669
X   0.008291067 0.010172375 0.005316099 0.007294589 0.003807516 0.002711993 0.00000431  0.004335257
X   0.00405819  0.006540146 0.002686623 0.004762546 0.00163977  0.003309515 0.00000789  0.001957311
X   0.0067902   0.010219971 0.004727677 0.0070145   0.002995965 0.005784826 0   0.003331767
X   0.005495986 0.008415487 0.003807673 0.00552374  0.002361458 0.005504102 0.000000805 0.002746723
X   0.005209265 0.008041805 0.003572746 0.005247395 0.002243573 0.005357235 0.0000016   0.002665558
X   0.007467104 0.010856364 0.004896178 0.008258422 0.003086705 0.00341522  0.00000285  0.002702047
Y   0.007670113 0.00940323  0.006406508 0.007372615 0.003627189 0.001986754 0.0000465   0.002585564
Y   0.004686043 0.00726605  0.003055345 0.005017251 0.002003367 0.003900496 0.000000476 0.002309116
Y   0.004177442 0.006972389 0.002680084 0.00474762  0.001885973 0.004097005 0.000000662 0.002295276
Y   0.004475911 0.006846188 0.003625686 0.004221453 0.002829967 0.002831975 0.00000357  0.001636054
Y   0.005848798 0.008090652 0.00340937  0.006071063 0.003104326 0.003234079 0.000000615 0.003722661
Y   0.006518662 0.009792935 0.004230156 0.007402922 0.002677034 0.002810624 0.000000494 0.002678868
Y   0.005428432 0.008254325 0.003481542 0.006025152 0.002306644 0.002731699 0.0000341   0.002559538
Y   0.006138387 0.007997736 0.003884782 0.005445355 0.003097547 0.003177233 0.0000197   0.003667534
Y   0.005268127 0.006967856 0.003245438 0.004784031 0.002785673 0.003147425 0.0000161   0.003360895
Y   0.006824798 0.008643139 0.00441859  0.006040691 0.003519989 0.003252144 0.0000123   0.003866648
Y   0.006756669 0.008882524 0.004407434 0.006087051 0.003316964 0.0038456   0.00000907  0.003794542
Y   0.006766808 0.008916508 0.004425926 0.006133224 0.003349647 0.00381514  0.0000122   0.003794175
Y   0.005288403 0.006397465 0.003993784 0.004615067 0.002988219 0.00219577  0.000224752 0.002179541
Y   0.007064576 0.010323695 0.00479257  0.007002344 0.003084699 0.005404821 0.00000601  0.003126432
Y   0.004992499 0.007830614 0.003199636 0.005100254 0.002196152 0.004706792 0.0000291   0.002574521
Y   0.007285894 0.009257734 0.004756352 0.006956334 0.003740338 0.002541691 0.0000017   0.003785755
Y   0.007284548 0.009060781 0.004695172 0.006907381 0.003923092 0.00237505  0.00000764  0.004050348
Y   0.00502541  0.006910243 0.00368707  0.004952021 0.003261628 0.002949114 0.000131343 0.002590292
Y   0.005605946 0.008788954 0.003741203 0.006023171 0.002442795 0.004465909 0.000000421 0.002661231
Y   0.003618486 0.006465109 0.002340945 0.004184876 0.001724263 0.003914174 0.00000106  0.002229038
Y   0.004858636 0.008062734 0.00315832  0.005736985 0.002185119 0.003723115 0.00000306  0.002509759
Z   0.005107876 0.008322495 0.003352967 0.005921168 0.002309924 0.003758071 0.00000206  0.002624568
Z   0.003509392 0.005521583 0.002485114 0.003173729 0.00126778  0.00355057  0.0000431   0.001887588
Z   0.003865434 0.005767412 0.002810001 0.003372551 0.001460574 0.003648029 0.0000457   0.001909999
Z   0.005177201 0.007826825 0.003958008 0.004716213 0.002546243 0.00398816  0.00000473  0.00242293
Z   0.006601694 0.009499528 0.004681797 0.00669723  0.003016417 0.004643877 0.00000777  0.003081804

Then, I used down below to create a figure as you can see in attachment (Figure1).

library(reshape2)
library(ggplot)
DvsO = read.table("Disease_vs_Outcome.txt", header = TRUE)
DvsO  <- data.frame(DvsO)
DvsO <- melt(DvsO, id.vars="Disease")


ggplot(DvsO, aes(Disease, value, fill=Disease)) + 
  geom_violin() + geom_boxplot(width=.5) + facet_wrap(~variable,scales="free") +  scale_fill_manual(values=c("#CC6666", "#00CC00","#3399FF"))

However, I'd like to generate a figure as this one (Figure2). Basically, I'd like to flip the figure into horizontal and put all the data together in one panel.

I'd be happy if you help me to figure out this? Thank you very much in advance.

Answer 1

You can flip the boxes with coord_flip (but you can't use free scales with that). I would advise to keep the facets, but use facet_grid instead. Making one panel is much harder. You can adjust the look of the facets if needed.

(You can probably do away with the y-axis in this plot.)

ggplot(DvsO, aes(Disease, value, fill=Disease)) + 
  geom_violin() + 
  geom_boxplot(width=.5) + 
  facet_grid(variable~.) +
  scale_fill_manual(values=c("#CC6666", "#00CC00","#3399FF")) +
  coord_flip() + 
  theme_bw() +
  theme(panel.margin = unit(0, "lines"))

EDIT:

And with points (you could also look at geom_dotplot):

ggplot(DvsO, aes(Disease, value, fill = Disease)) + 
  geom_violin(color = NA, alpha = 0.8) + 
  geom_point(position = position_jitter(w = 0.5, h = 0), alpha = 0.7) +
  facet_grid(variable~.) +
  scale_fill_manual(values=c("#CC6666", "#00CC00","#3399FF")) +
  coord_flip() + 
  theme_bw() +
  theme(panel.margin = unit(0, "lines"), 
        axis.text.y = element_blank(),
        axis.ticks.y = element_blank(),
        axis.title.y = element_blank())