CODEWITHSUNDEEP
ggplot2
cagias
cagias

Reputation: 39

Missing values in barplot

With the help I already got here on SO, I was able to built the barplot I wanted.

Now I'd like to do exact the same plot using new data that is very similar to the previously one, although it's not working and I can't figure out why.

The data is as it follows: 

 > dput(pcwdegs)
structure(list(comp = c("C0vsC12", "C0vsC12", "C0vsC24", "C0vsC24", 
"C0vsT12", "C0vsT12", "C0vsT24", "C0vsT24", "C12vsC24", "C12vsC24", 
"C12vsT12", "C12vsT12", "C12vsT24", "C12vsT24", "C24vsT12", "C24vsT12", 
"C24vsT24", "C24vsT24", "T12vsT24", "T12vsT24"), reg = c("up", 
"down", "up", "down", "up", "down", "up", "down", "up", "down", 
"up", "down", "up", "down", "up", "down", "up", "down", "up", 
"down"), count = c(7L, 15L, 11L, 5L, 51L, 50L, 49L, 38L, 10L, 
2L, 54L, 44L, 55L, 32L, 24L, 32L, 12L, 17L, 31L, 22L)), class = "data.frame", row.names = c(NA, 
-20L))

The code and warning messages I'm getting (saying ALL observations were removed) and the plot:

ggplot(data = pcwdegs, 
       mapping = aes(x = comp, fill = reg,
                     y = ifelse(test = reg == "down", 
                                yes = -count, no = count))) +
  geom_bar(stat = "identity") +
  scale_y_continuous(labels = abs, limits = max(pcwdegs$count) * c(-1.05,1.05)) +
  geom_text(aes(label=count), vjust=0.5, color="black", size=3.0, nudge_y = c(-1.5,1.5))+
  labs(y = "DEGs related to plant cell wall", x = "Groups comparisons") + 
  scale_fill_manual(values=c("#98FB98","#FA8072","#00FF00","#FF0000"), name = "Expression regulation", labels = c("down-regulated", "up-regulated")) +
  scale_x_discrete(limits=c("T12xT24", "C24xT24", "C24xT12", "C12xT24", "C12xT12", "C12xC24", "C0xT24", "C0xT12", "C0xC24","C0xC12")) +
  coord_flip()    

Warning messages:
    1: Removed 20 rows containing missing values (position_stack). 
    2: Removed 20 rows containing missing values (geom_text). 
    3: Position guide is perpendicular to the intended axis. Did you mean to specify a different guide `position`?

enter image description here

Looking online for solutions, the suggestions were to check if there were NA values (there weren't) and to set higher limits in limits = max(pcwdegs$count) * c(-1.05,1.05) (as it was already suggested on my first question and done). So I still have no clue why such similar data is behaving different to the code.

Upvotes: 0

Views: 758

Answers (1)

stefan
stefan

Reputation: 123963

The problem arises because the categories in your new dataset don't fit into the limits defined by scale_x_discrete. The categories in your new df have an vs in the middle, while in teh limits it's a x. Simply replace x with vs in the limits and everything is fine.

library(ggplot2)

ggplot(data = pcwdegs, 
       mapping = aes(x = comp, fill = reg,
                     y = ifelse(test = reg == "down", 
                                yes = -count, no = count))) +
  geom_bar(stat = "identity") +
  scale_y_continuous(labels = abs, limits = max(pcwdegs$count) * c(-1.05,1.05)) +
  geom_text(aes(label=count), vjust=0.5, color="black", size=3.0, nudge_y = c(-1.5,1.5))+
  labs(y = "DEGs related to plant cell wall", x = "Groups comparisons") + 
  scale_fill_manual(values=c("#98FB98","#FA8072","#00FF00","#FF0000"), name = "Expression regulation", labels = c("down-regulated", "up-regulated")) +
  #scale_x_discrete(limits=c("T12xT24", "C24xT24", "C24xT12", "C12xT24", "C12xT12", "C12xC24", "C0xT24", "C0xT12", "C0xC24","C0xC12")) +
  scale_x_discrete(limits=c("T12vsT24", "C24vsT24", "C24vsT12", "C12vsT24", "C12vsT12", "C12vsC24", "C0vsT24", "C0vsT12", "C0vsC24","C0vsC12")) +
  coord_flip()

Created on 2020-05-23 by the reprex package (v0.3.0)

Upvotes: 1

Related Questions