PDA

View Full Version : PLINK and 23andme raw data file



sonofaroli
02-25-2017, 08:08 AM
Is there a way using Plink to convert 23andme raw data text files into a .ped files? Or is there other software that does this? How is this generally done? Thanks

Chad Rohlfsen
02-25-2017, 08:28 AM
rename your 23andme file to File.txt

Next you enter the following example line in your terminal

/home/chad/plink/plink --23file File.txt --make-bed --out Chad

Do you know how to extract only certain markers when you merge sets?

After you merge it with your set, you use convertf to go to ped and geno. I'm not sure of any other way.

You have Admixtools, correct?

Generalissimo
02-25-2017, 08:48 AM
Also manually delete all the lines with X in the 23andMe text file, or rename the X to 23, because PLINK won't take the X and you'll see an error.

anglesqueville
02-25-2017, 02:29 PM
Not completely off topic: did anyone ever use aConv ( from Felix C.) to convert a 23&me or ftdna data file in eigenstrat format?

sonofaroli
02-25-2017, 06:56 PM
Thanks for the prompt reply. I didn't expect such a fast response. So thanks for that. I'm away from my laptop at the moment so will put your suggestions into action when I get back home. I haven't used AdmixTools but I have been playing around with Admixture over the last few days using a public dataset, trying to determine the CV error value for various K. New to Linux and only just starting to experiment with these tools. I'm looking ahead to when I will want to merge my data, hence the question. Actually, your answer takes me to the .bed file which is ultimately what I want. I was going to use the .ped file to get to the.bed using the standard plink commands shown in the manuals but didn't know how to get to the initial .ped step.

sonofaroli
02-25-2017, 08:45 PM
Thanks Generalissimo. I'll take care to follow that suggestion. Presumably I also need to delete the introductory text at the head of the file as well.