plink mergelist put many individuals into same binary file set

Question

[gulcek]

We have some difficulties about using 23andMe data on plink. We have raw 23andMe snp data for each individual (total of 756). We converted each individual file to binary fileset .bed,.bim, .fam via plink.

plink --23file $file --out $a

After creating binary set for each individual, we converted them into .map and .bed format.

plink --bfile binary_filesets --a2-allele ../All.vcf 4 3 '#' --recode --out $a

Then, we want to collect info of all individuals into same .bed file( or binary fileset.)

plink --file genom0 --merge-list ../filesets3.txt --make-bed --out ../combined/combdata

We assume --merge-list does this job, however, it does overwrite one individual into another. It doesn't create many individuals x info (matrix-like) data. How could we obtain one combined binary fileset for many individuals?

[Dawn]

Hello! I am also new to Plink. I encountered the same problem. How did you solve it?

[ATpoint]

Did you follow the answer of the developer chrchang523 below? Also please leave the answer field for actual answers and use the comment buttom instead.

[Dawn]

Yes, I want to follow the answer of the developer chrchang523. But since I am very new to Plink, I don't know how to assign a different ID to each sample. Can you give me some instructions? Thank you.

[Kevin Blighe]

I don't know how to assign a different ID to each sample.

This intention was not stated in your original question. Please try to explain again exactly what you want to do. State:

• what data you have
• what you want to do
• what you have already tried and the error messages

[Kevin Blighe]

New question: How to merge 23andme samples in Plink? Each .ped/.map file represent a sample.

Answer 1

[chrchang523]

1. --a2-allele is pointless with --recode, since the .ped/.map format does not track reference alleles. You must use --make-bed instead of --recode.
2. You need to assign a different ID to each sample (this is easy to do during the --23file step). Otherwise --merge-list assumes all the genotype calls are supposed to be for the same sample.