Question about Mendel error results after swapping parent/offspring in FAM file

[jjjhbj ghhh]

Hi all

i am a plink1.9 new user,when i using PLINK1.9 to perform Mendelian error checks on trios, and I encountered a situation which I would like to confirm with you.

Originally, my .fam file for one nuclear family was:
1 213 0   0   1 1
1 214 0   0   2 1
1 215 213 214 0 2
Here, 213 and 214 are the founders (father and mother), and 215 is the offspring (parents = 213, 214).
By mistake, I modified the .fam file to:
1 213 0   0   1 1
1 215 0   0   2 1
1 214 213 215 0 2
In this incorrect version, 213 and 215 are treated as founders, and 214 is treated as the offspring (parents = 213, 215).
Conceptually, the true family relationship is:
Correct structure: 213 & 214 → 215 (215 is the child)
But in the modified FAM file, PLINK sees:
Altered structure: 213 & 215 → 214 (214 is the child)
I expected that this change in the FAM structure might alter the Mendelian error results, since the parent–offspring roles are swapped. However, when I ran PLINK’s Mendelian error check (--mendel) on both versions of the FAM file (keeping the genotypes unchanged), I found that:
The total number of Mendelian errors was exactly the same, and
The specific SNPs flagged as Mendelian errors were also exactly the same.
I repeated this procedure for 24 nuclear families (performing similar swaps of parent/offspring lines in the .fam file), and in all cases the Mendelian error counts and error SNP sets were identical between the original and the swapped FAM files.

my question is :

From the implementation perspective, is it expected/possible that swapping the parent and offspring roles in the .fam file (as in the example above) would still produce identical Mendelian error results?

[Chris Chang]

0. In general, you should post a full .log file when asking a question like this.

1. In this case, the result was the same because you did not actually change the pedigree structure, you only *renamed* the samples.  The first line of the .fam corresponds to the first sample in the .bed, etc.

If you had instead edited to

1 213 0   0   1 1

1 214 213 215 0 2
1 215 0   0   2 1

you would have seen different results.

--
You received this message because you are subscribed to the Google Groups "plink2-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to plink2-users...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/plink2-users/f4e76a71-99b2-4ea0-89da-aca770c984b7n%40googlegroups.com.