How do I merge imputed GWAS data

Question

[reagan]

I have a cohort that part of the participants were genotyped using the Illumina 2.5M DNA microarray chip and another group of participants using H3Africa chip. I have imputed them using HRC server. To increase the power for my association analyses, I would like to merge them before I conduct the GWAS rather than doing a meta-analysis after the GWAS. I have read that I should only merge intersecting (common in both datasets) SNP. I know how to do this for unimputed data using PLINK2, but I don't know how to do this in imputed data with several chromosomes. Does some have any lead on this? Any help would be greatly appreciated.

[4galaxy77]

There's no fundamental difference between imputed and non-imputed data in terms of merging. I would say that it is likely you only want to keep positions that have been genotyped (common SNPs). plink2 sounds like a good way to merge them - what format are both the datasets in?

[Đạt]

Hello, Could you please share your experience of how to do analysis association with the imputed genotype after we impute the data? Someone say they combine imputed data and non-imputed data to one data and filter imputed SNP with low quality but other use different package for genetic association of imputed data and non-imputed data. How do you do in this case? thank you