plotting graph in R from a bed file

[aevgup]

Hi Everyone,

I have a bed file. I am showing you a section of the bed file.

Slc19a1	81	144	1
Slc19a1	192	255	6
Slc19a1	273	336	7
Slc19a1	363	426	4
Slc19a1	465	528	3
Slc19a1	540	603	5
Slc19a1	813	876	5
Slc19a1	912	975	6
Slc19a1	987	1050	8
Slc19a1	1062	1125	6
Slc19a1	1170	1233	8
Slc19a1	1278	1341	6

The 4th column is the number of reads for particular interval. For eg the read count is 6 for the interval 192-255.

I want to draw a histogram in R for these intervals with x axis having the intervals and y axis the counts

How can I do it. I am a beginner in R.

Thanks for the help

Regards

[ECO]

There will be a lot of fiddling to get it to look better, but here is my approach.

1. Get the data into a dataframe.

I use this function to read from the clipboard on a mac. You can also use read.table to read the whole bed file more efficiently:

>cboardf<- function(){
>read.table(pipe("pbpaste"),sep='\t',header=F)}

> d <- cboardf()

> str(d)
'data.frame':	12 obs. of  4 variables:
 $ V1: Factor w/ 1 level "Slc19a1": 1 1 1 1 1 1 1 1 1 1 ...
 $ V2: int  81 192 273 363 465 540 813 912 987 1062 ...
 $ V3: int  144 255 336 426 528 603 876 975 1050 1125 ...
 $ V4: int  1 6 7 4 3 5 5 6 8 6 ...

2. Then you need to make a "name" for your regions...just a string...

> d$V5 <- paste(d$V2,"-",d$V3)  #make a new string for the region in column called V5

> str(d)
'data.frame':	12 obs. of  5 variables:
 $ V1: Factor w/ 1 level "Slc19a1": 1 1 1 1 1 1 1 1 1 1 ...
 $ V2: int  81 192 273 363 465 540 813 912 987 1062 ...
 $ V3: int  144 255 336 426 528 603 876 975 1050 1125 ...
 $ V4: int  1 6 7 4 3 5 5 6 8 6 ...
 $ V5: chr  "81 - 144" "192 - 255" "273 - 336" "363 - 426" ...

3. Then make a plot (using ggplot2)

qplot(V5,V4,data=d,geom='bar') + opts(axis.text.x=theme_text(angle=90))

Voila:


You'll have to work a lot harder to make it look better, label axes is easy, sorting the X regions is more annoying, but pretty easy.

Good luck.