Question about family ID

[Daiki Tagami]

Hi all,

I'm a new PLINK user, and I'm trying to conduct GWAS in PLINK.

My genetic dataset is generated from simulations and it does not have pedigree information, so I'm planning to use the `--double-id` tag to set the family ID as the individual ID.

I'm wondering about the usage of family ID in PLINK, and am I correct with saying that PLINK only uses family ID in family based association test?

Thank you for your help.

[Christopher Chang]

- I recommend "--const-fid 0", which is PLINK 2.0's usual way of handling single-part sample IDs, over --double-id.

- Family IDs also affect Mendel-error clearing/filtering, but yes, PLINK is perfectly usable without them.

[Daiki Tagami]

Thank you very much for your help. It was very clear to understand.