Plot line graph when counts are not automatically given

Question

[kristina.mahan]

I want to plot Depth of Coverage values in R. I've extracted all the Coverage values from my VCF and want to plot it. The x-axis would be Depth and the y-axis would be counts. I don't have the counts specified I just have a .txt file with a column of values.

In R:

library(ggplot2)
filename <- "KMM1_raw_variants_DP_values_10102018.txt"
my_data <- read.csv(filename, sep="\t", header=FALSE)
head(my_data)

   V1
1 350
2 432
3 431
4 479
5 469
6 410
  

names(my_data)[1] <- c("Coverage")

  Coverage
1      350
2      432
3      431
4      479
5      469
6      410
  

ggplot(my_data,aes(x=Coverage, y=counts)) + geom_line()

Error in FUN(X[[i]], ...) : object 'counts' not found
  

How do you specify that you want R to count the times each coverage is present so that I can see the Depth of Coverage so I can know what to filter out when using GATK? First graph here: http://mbontrager.org/blog/2016/08/17/Variant-Exploration

Thanks!!

Answer 1

[Friederike]

You don't want a line plot, you want a histogram/density plot where you capture how many times a given number appears in your coverage column.

In your case you could try:

P <- ggplot(my_data, aes(x = Coverage)) + geom_density()
## zooming in
P + coord_cart(xlim = c(0, 250))

## histogram
ggplot(my_data, aes(x = Coverage)) + geom_histogram()

[kristina.mahan]

When I do that I get this error:

Error: StatBin requires a continuous x variable: the x variable is discrete. Perhaps you want stat="count"?

[Friederike]

are you sure that the entries of my_data$Coverage are numbers? what does str(my_data) return?

[kristina.mahan]

'data.frame': 725952 obs. of 1 variable: $ V1: num 478 569 568 620 609 545 242 240 229 346 ...