subset x axis when using facet

Possibly this question has been asked before but I don't manage to find a solution. I have a TSV (test.tsv) that looks like this:

TPM variants    NOI_ID  HLA NAL
324.28  several p1  one 2
169.21  NA  p1  two 15
154.78  NA  p1  three   0
143.31  NA  p1  four    2
468.7   NA  p2  five    0
322.76  several p2  six 2
620.98  NA  p2  two 0
591.17  NA  p2  seven   0
637.74  NA  p3  eight   4
519.8   NA  p3  nine    10
1439.58 NA  p3  ten 23
122.05  NA  p4  five    14
149.74  NA  p4  eleven  77
213.7   NA  p4  twelve  100
162.53  NA  p4  one 15

Using this code:

fig13 <- read.table("test.tsv", sep="\t", header=T, check.names=FALSE)
fig13 = fig13[-which(grepl("_w8", fig13$NOI_ID)),]
fig13_melt <- melt(fig13[c("TPM", "NOI_ID", "HLA", "variants")], id=c("NOI_ID", "HLA", "variants"))

hla_barplot = ggplot(fig13_melt, aes(x=1:nrow(fig13_melt), y=value, fill=variants))  
     geom_bar(stat="identity", position="dodge")   ggtitle("HLA status", subtitle = "all samples")   theme_bw()  
     theme(axis.title.y=element_blank(), axis.title.x=element_blank(), 
           plot.title = element_text(face = "bold", size = 15, hjust = 0.5),
           plot.subtitle=element_text(size=12, hjust=0.5, face="italic"),
           axis.text.y = element_text(size=6),
           axis.text.x = element_text(angle = 90, hjust=1, size=6))  
     geom_text(aes(label = round(value,0), vjust = 1, size = 1))  
     scale_x_discrete(labels=fig13_melt$HLA, breaks=1:nrow(fig13_melt), limits=factor(1:nrow(fig13_melt)), name='HLA')
hla_barplot

I can plot the following:

However when I try to use facet_wrap to split the plot:

hla_barplot   facet_wrap(.~ NOI_ID, nrow = 1)

I get the following:

So, as far as I can see, each subplot has all the x labels of the whole dataset. Is there a way to just plot in the x-axis the labels of each subgroup?

CodePudding user response：

The issue is that you mapped 1:nrow(...) on x and fixed the limits of the x scale limits=factor(1:nrow(fig13_melt)). Doing so scales="free_x" as suggested in the comments will not work.

Instead I would suggest to map HLA on x, get rid of the scale_x_discrete and free your scales using scales="free_x".

If you want to reorder your bar chart I would suggest to set the factor levels of your HLA column accordingly. To this end I add a helper column HLA2 to your df to which I add the row number so that your two "one" categories get unique. Afterwards I use a function for the labels argument of scale_x_discrete to remove the row numbers:

library(ggplot2)

fig13_melt$HLA2 <- paste0(fig13_melt$HLA, seq(nrow(fig13_melt)))
fig13_melt$HLA2 <- forcats::fct_inorder(fig13_melt$HLA2)

ggplot(fig13_melt, aes(x = HLA2, y = value, fill = variants))  
  geom_bar(stat = "identity", position = "dodge")  
  geom_text(aes(label = round(value, 0), vjust = 1, size = 1), show.legend = FALSE)  
  labs(title = "HLA status", subtitle = "all samples", x = "HLA")  
  theme_bw()  
  theme(
    axis.title.y = element_blank(), axis.title.x = element_blank(),
    plot.title = element_text(face = "bold", size = 15, hjust = 0.5),
    plot.subtitle = element_text(size = 12, hjust = 0.5, face = "italic"),
    axis.text.y = element_text(size = 6),
    axis.text.x = element_text(angle = 90, hjust = 1, size = 6)
  )  
  scale_x_discrete(labels = ~ gsub("\\d $", "", .x))  
  facet_wrap(~NOI_ID, nrow = 1, scales = "free_x")

DATA

fig13_melt <- structure(list(NOI_ID = c(
  "p1", "p1", "p1", "p1", "p2", "p2",
  "p2", "p2", "p3", "p3", "p3", "p4", "p4", "p4", "p4"
), HLA = c(
  "one",
  "two", "three", "four", "five", "six", "two", "seven", "eight",
  "nine", "ten", "five", "eleven", "twelve", "one"
), variants = c(
  "several",
  NA, NA, NA, NA, "several", NA, NA, NA, NA, NA, NA, NA, NA, NA
), variable = structure(c(
  1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
  1L, 1L, 1L, 1L, 1L, 1L
), levels = "TPM", class = "factor"), value = c(
  324.28,
  169.21, 154.78, 143.31, 468.7, 322.76, 620.98, 591.17, 637.74,
  519.8, 1439.58, 122.05, 149.74, 213.7, 162.53
)), row.names = c(
  NA,
  -15L
), class = "data.frame")

CodePudding user response：

There are 3 parts to be revised:

Use x = factor(1:nrow(fig13_melt)) in aes() to make the x-axis discrete instead of continuous. In this way, you can also change the label order of x-axis by setting levels=... in factor().

ggplot(fig13_melt, aes(x = factor(1:nrow(fig13_melt)), y = value, fill = variants))

Remove limits = factor(1:nrow(fig13_melt)) in scale_x_discrete(); Otherwise, all of the categories in the x-axis will be shown.

scale_x_discrete(labels = fig13_melt$HLA,
                 breaks = factor(1:nrow(fig13_melt)),
                 name='HLA')

Add scales = "free_x" into facet_wrap():

facet_wrap(.~ NOI_ID, nrow = 1, scales = "free_x")

Data

fig13_melt <- structure(list(NOI_ID = c("p1", "p1", "p1", "p1", "p2", "p2",
"p2", "p2", "p3", "p3", "p3", "p4", "p4", "p4", "p4"), HLA = c("one",
"two", "three", "four", "five", "six", "two", "seven", "eight", 
"nine", "ten", "five", "eleven", "twelve", "one"), variants = c("several",
NA, NA, NA, NA, "several", NA, NA, NA, NA, NA, NA, NA, NA, NA
), variable = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L), levels = "TPM", class = "factor"), value = c(324.28,
169.21, 154.78, 143.31, 468.7, 322.76, 620.98, 591.17, 637.74,
519.8, 1439.58, 122.05, 149.74, 213.7, 162.53)), row.names = c(NA,
-15L), class = "data.frame")