Entering edit mode
4.3 years ago
Dawn
•
0
I got some 23andme samples and want to analysis them. First I converted these 23andme files into .ped files. Then I merge these .ped files to one file in plink, but I got errors. By looking up the information, I found that I need to assign a different ID to each sample. But I don't know how to do it? Can someone tell me?
I have added the
plinktag to your question. Also, to help, please edit your question to show the commands that you have used and the error messages received.Finally, I presume that you have already read this: https://www.cog-genomics.org/plink/1.9/data#update_indiv
Here is some information about what I have done so far:
I convert our 23andme files to .bed format with the following command:
plink --23file 1.txt Person1 AGENT M --out 1
plink --23file 2.txt Person2 AGENT M --out 2
plink --23file 3.txt Person3 AGENT M --out 3
so on...
This seems to work fine. Then I merge with other files similarly created using the following command:
plink --bfile 1--merge-list list.txt --make-bed --out CombinedPerson
The file list.txt is formatted as follows for simplification:
2.bed 2.bim 2.fam
3.bed 3.bim 3.fam
4.bed 4.bim 4.fam
so on....
This also seems to work, but the software I put the CombinedPerson file through leads me to believe a small mess was created.
Warning: Multiple positions seen for variant 'rs4477212'.
Warning: Multiple positions seen for variant 'rs3094315'.
Warning: Multiple positions seen for variant 'rs3131972'.
1920500 more multiple-position warnings: see log file.
Please edit and add these commands to your original post.
Please use the formatting bar (especially the
codeoption) to present your post better when you edit.Thank you!
It looks like you have 23andMe files across several years, using different reference genome builds.
The top of each 23andMe file should mention the reference genome version. You can merge within each version first, perform a “lift-over” operation to convert to one set of coordinates, and then perform a final merge.