Transforming Plink Files

Question

[FFK534]

Does anyone have any suggestions for combining .ped & .map files from Plink and transforming them into a different format? For example, the peds look like this:

#FID      IID        PAT     MAT     SEX     STATUS     G1     G2     G3     G4     G5     G6     G7     G8     G9     G10
12322     12322A     0       0       1       1          1      1      1      1      1      1      1      1      2      1
12322     12322B     0       0       2       0          1      1      1      1      2      2      2      1      1      2
12322     12322C     0       0       2       1          2      1      1      1      1      1      1      1      1      1

and the maps look like this:

#CHR     G                   GD     BP
1        1_135195_A/G        0      135195
1        1_135203_G/A        0      135203
1        1_136596_GGGG/-     0      136596
1        1_136604_G/C        0      136604
1        1_136619_G/A        0      136619
1        1_136620_C/T        0      136620
1        1_136635_T/G        0      136635
1        1_136645_G/-        0      136645
1        1_136652_A/G        0      136652
1        1_136779_G/A        0      136779

And what I'd like is this:

1_135195_A/G     1_135203_G/A     1_136596_GGGG/-     1_136604_G/C     1_136619_G/A     1_136620_C/T     1_136635_T/G     1_136645_G/-     1_136652_A/G     1_136779_G/A     STATUS
1                1                1                   1                1                1                1                1                2                1                1
1                1                1                   1                2                2                2                1                1                2                0
2                1                1                   1                1                1                1                1                1                1                1

Where the 2nd column in the 2nd file becomes the header of the third file and the 6th column of the 1st file becomes the final column of the 3rd file.

Answer 1

[Ram]

If you're willing to use R, read these to data frames, extract as vectors and restructure as you see fit! Also, please only use relevant tags - I don't see how SQL is relevant here.

[FFK534]

I should note there are thousands of columns in the first file and thousands of rows in the second file. How can I extract multiple vectors without listing the columns/rows individually?

[Ram]

You can transpose and slice. Speaking of, try Python - that might make it a bit more flexible, but you'l have to spend more time on the logic.

Answer 2

[FFK534]

This works:

ped <- read.table ("/pathtodata.ped")
map <- read.table ("/pathtodata.map")

map$V1 <- map$V3 <- map$V4 <- NULL
tmap <- t(map)
ped$V1 <- ped$V2 <- ped$V3 <- ped$V4 <- ped$V5 <- NULL
status <- ped$V6
ped$V6 <- NULL
colnames(ped) <- c(tmap)
ped$Status <- status

Answer 3

[zx8754]

Look into --recodeA option in plink. That would create single *.raw format. http://pngu.mgh.harvard.edu/~purcell/plink/dataman.shtml#recode