Genotype frequency per sample

Question

[HG]

Hi everyone, I have a plink file and I want to get the distribution of calls (AA/Aa/aa) for each sample. How can I get to this? Any help would be appreciated.

[finswimmer]

Hello,

could you please post an example of your file?

fin swimmer

[HG]

Hi, Plink files (my input files) are .map and .ped files:

.ped file is as follows:

A text file with the following fields:

1. Family ID
2. Individual ID
3. calls for each SNP

.map file:

A text file with no header file, and one line per variant with the following 3-4 fields:

1. Chromosome code.
2. Variant identifier
3. Position in morgans or centimorgans (optional; also safe to use dummy value of '0')
4. Base-pair coordinate

Answer 1

[chrchang523]

"plink --file ... --het" will give you the number of Aa calls for each sample.

If you also need AA/aa, you'd need to define which alleles are A and which ones are a. Once you have, you can merge in a synthetic sample with all aa calls, use --merge to merge that sample with your real dataset, and then run "--genome full". Then look at the lines of the .genome file which include your artificial all-aa sample; the IBS0 and IBS2 columns of those lines will give you the additional counts you need.

[HG]

Thanks for your comments. Yes, I need AA/Aa/aa. --het will return a text file with the following columns: 1. FamilyID, 2.Within-family ID, 3. Observed number of homozygotes, 4. Expected number of homozygotes, 5. Number of non-missing autosomal genotypes, 6. Method-of-moments F coefficient estimate. No information about heterozygotes (Aa)! Also, how can I perform what you explained in the second part of your comments? Any scripts in R? Thanks