How to replace row with another row in linux having NA and nucleotide bases

Question

[Kritika]

I have file 1:

A1  1  NA
A1  2  NA
A1  3  NA
A1  4  A
A1  5  G
A1  6  T
A1  7  NA
A1  8  NA
A1  9  NA
A2  1  NA
A2  2  NA
A2  3  T
A2  4  NA

And file 2:

A1  4  A
A1  5  C
A1  6  T
A2  3  T

I want to replace row number 4,5,6 and 3 of A2 from file 1 with a value of 4,5,6 and 3 from file 2

Expected Output in new file3:-

 A1  1  NA
 A1  2  NA
 A1  3  NA
 A1  4  A
 A1  5  C
 A1  6  T
 A1  7  NA
 A1  8  NA
 A1  9  NA
 A2  1  NA
 A2  2  NA
 A2  3  T
 A2  4  NA

I want to try this in Linux. I tried googling but I could not find better answers.

I tried in linux.

diff file2 file1
join -a 2 file2 file1 | cut -d ' ' -f -2

But i am not getting answers

Answer 1

[Paul]

you can use awk:

cat file2 file1 | awk '!seen[$1,$2]++' | sort -k 1V,1 -k 2n,2 > result

Output:

A1  1  NA
A1  2  NA
A1  3  NA
A1  4  A
A1  5  C
A1  6  T
A1  7  NA
A1  8  NA
A1  9  NA
A2  1  NA
A2  2  NA
A2  3  T
A2  4  NA

[Kritika]

No its not working it is not making any file ! No output

[Paul]

Hi, Are your files tab delimited? cat -T input is what result?

[Kritika]

its giving me what is there in file value of file1

[Paul]

if you are try to print first column - awk '{print $1}' input - does it work? When I test my on-liner on your pasted data above it is works.

[Kritika]

Yes its printing the first column

[Kritika]

The above code is not making any file

[Kritika]

The columns are seperated by 2 spaces