Hii all
This may be a very naive question but in standard GWAS QC, we usually remove variants that significantly deviate from Hardy-Weinberg equilibrium. However, for rare variant analyses like in SAIGE, HWE tests may be less reliable because of low allele counts. In practice, do we still apply the same HWE filtering criteria to rare variants, or are different QC thresholds recommended? How much can this affect downstream association results?
Hi! Using different QC thresholds for rare variants are not recommended. For rare variants, you lack the statistical power to detect meaningful deviations from HWE. If you were to ignore HWE or place a threshold on rare variants, you could produce false positive results.
Thank you for your response and explanation.