QC exclusion of A/T or G/C alleles to avoid strand issues

Question

[_quantum_girl_]

I read in one paper they excluded (SNVs) with A/T or G/C alleles to avoid strand issues. What does this even mean? Why would there be "strand issues"?

Thanks for your help!

[bk11]

This could be because they are probably merging two or more datasets. If you are merging two or more datasets, it will be very important that SNPs sets are concordant in terms of positive or negative strands among the datasets. If not more that two alleles will be observed in the merged dataset.

[_quantum_girl_]

They aren't merging any datasets. Just for reference the paper is this: https://www.nature.com/articles/s41380-022-01674-9#MOESM1. Citing: "Quality control was carried out with PLINK v2.0 using the following criteria: [...] exclusion of single nucleotide variants (SNVs) with call rate <95%, Hardy-Weinberg equation (HWE)<1e-4, minor allele frequency (MAF)<0.01, and with A/T or G/C alleles to avoid strand issues".