Using SAIGEgds for non additive models

[Richard-Packer]

Hi,

I use SAIGEgds for smaller scale analyses and think it's a great tool (SNP numbers approximately 1000). Iwas wondering if SAIGEgds could be used for alternative models of inheritance. So, recoding the dosage to reflect a dominant model. That also got me thinking if it could also be used for genetic risk scores, in effect having a single "SNP" that reflected the risk score. I can see how this would be achieved using the gds structure and recoding but just wondered if you had any insight into whether this approach invalidated the underlying principles of the analysis.

Best,
Richard