[R] [Q] It it possible to create the data frame only non-zero data column?

[Hiroyuki Sato]

Hello all.

I re-post this question by e-mail.
(I posted via google-group. But It's not posted yet.)

I'm newbie GNU R.

I would like to compare two datas.
How to select columns which has non-zero datas?.
It it possible to create the data frame only VAL3(non-zero data) column
with command?

Formatted sample.
https://gist.github.com/hiroyuki-sato/cb36584f6cd5845b6c3e

sample1.txt

  ID,VAL1,VAL2,VAL3
  ID1,0,2,3
  ID2,0,2,3
  ID3,0,2,3

  real data has 5000 columns.

sample2.txt

  ID,VAL1,VAL2,VAL3
  ID1,0,2,3
  ID2,0,2,3
  ID3,0,2,2

  The difference sample1 and sample2 is ID3/VAL3.
    sample1: 3
    sample2: 2

R commands.

  sample1 <- read.table("sample1.txt",header=T,sep=',')
  sample2 <- read.table("sample2.txt",header=T,sep=',')

  result <- sample1[,2:4] - sample2[,2:4]
  result
    VAL1 VAL2 VAL3
  1    0    0    0
  2    0    0    0
  3    0    0    1

I would like to create data frame which has non-zero value columns.
Could you tell me how to do it?

Best regards.

--
Hiroyuki Sato.

	[[alternative HTML version deleted]]