A question about the missing or not observed alleles in PLINK datasets

Question

[abedkurdi10]

Hello everyone,

I would like to ask a question about the SNP array: I have eight PLINK datasets and I noticed I have quite a large number of variants with missing or no observed allele. What could be the cause of having a large number of variants with no observed allele?

Much appreciated!

[bk11]

A large number of variants with no observed allele could be due to genetic drift in which allele frequencies of a population change over generations due to chance. Another possible reason could be that these variants are rare and are often not observed in a sample due to their low frequency.

[abedkurdi10]

Hello bk11 again,

I noticed that all the homozygote to reference allele variants have "missing" alternative allele. What should we do in such case, especially if we are thinking of merging multiple datasets, remove them? I noticed also that excluding them increases the genotyping rate. Any thoughts here?

I appreciate any information and guidance regarding this topic.