I have the following data
ID Affection rs3117294 rs2747453 rs2747454 rs2747457 rs3131888
D0024949 0 C_C A_G A_G A_A A_G
D0024302 0 A_C A_A A_G A_A A_A
D0023151 0 C_C A_G A_A A_A G_G
D0022042 0 A_C A_A G_G A_A A_A
D0021275 0 C_C A_G A_G A_A A_G
D0021163 0 A_A A_A G_G A_A A_A
D0020795 0 A_A A_A G_G A_C A_G
D0020691 0 A_A A_A G_G A_C A_G
D0019121 0 A_A A_A G_G C_C G_G
and I want to creat a .ped file for plink how can I do that?
Thanks
•
written
by
mms140130
2 answers
This is possible but you are missing a lot of information, namely:
- Family ID (FID)
- Paternal ID (PID)
- Maternal ID (MID)
- gender/sex
You are also missing a map file. See the map file format here.
You can create a temporary (and incomplete) map with the following code, which is specific for your dataset:
head -1 plink.raw | sed 's/ \+/\n/g' | sed '1,2d' | awk '{print "0\t"$0"\t0\t0"}' > plink.map
cat plink.map
0 rs3117294 0 0
0 rs2747453 0 0
0 rs2747454 0 0
0 rs2747457 0 0
0 rs3131888 0 0
Next, you have to edit your main data to get it into a pseudo-PED format. Read about PED files here, and their input here.
sed '1d' plink.raw | sed 's/_/ /g' > plinkv2.raw
cat plinkv2.raw
D0024949 0 C C A G A G A A A G
D0024302 0 A C A A A G A A A A
D0023151 0 C C A G A A A A G G
D0022042 0 A C A A G G A A A A
D0021275 0 C C A G A G A A A G
D0021163 0 A A A A G G A A A A
D0020795 0 A A A A G G A C A G
D0020691 0 A A A A G G A C A G
D0019121 0 A A A A G G C C G G
Then you can input your data, but you have to specify that you're missing FID, PID, MID, and gender/sex.
/Programs/plink1.90/plink --file --ped plinkv2.raw --map plink.map --no-fid --no-sex --no-parents
PLINK v1.90b3.38 64-bit (7 Jun 2016) https://www.cog-genomics.org/plink2
(C) 2005-2016 Shaun Purcell, Christopher Chang GNU General Public License v3
Logging to plink.log.
Options in effect:
--file
--map plink.map
--no-fid
--no-parents
--no-sex
--ped plinkv2.raw
15037 MB RAM detected; reserving 7518 MB for main workspace.
.ped scan complete (for binary autoconversion).
Performing single-pass .bed write (5 variants, 9 people).
--file: plink.bed + plink.bim + plink.fam written.
Are you using MAC? I use linux (Ubuntu).
All you need for the map is a list of your SNP IDs, surround by 1 column of zeros on the left, and 2 columns of zeros on the right.
What is the aim of your analysis, by the way? - are you just doing association testing?
Looks like your 'PED' file is incomplete. PLINK found 199 variants in your map, and therefore expected 2 * 199 genotypes in the PED file, plus 2 extra columns for sample ID and phenotype.
Just look over the files to ensure that there are no formatting issues. If you only have 199 SNPs, this should not take long
My code is only based on the small sample that you provided, which may not be applicable to the entire dataset.
One more thing that you could try is opening your file with the vi editor and checking to see if there is a ^M at the end of each line. In that case, use the dos2unix command to get rid of these, and then retry the code.
plink --bfile input --recode --tab --out output
Log in to answer this question.