NA is not allowed in genotype, use 'MVP.Data.impute' to impute

[suvihon]

I tried using rMVP to do GWAS with my data, I must say I'm impressed by the speed and figures it makes as a standard output. Others in my lab tend to run GWAS without imputing missing data. Seems like rMVP doesn't have this option so I'm a bit unsure what to do.

Based on the documentation I understood it is not recommended to do imputation in rMVP, but I should impute my data using something else (like BEAGLE) and then use rMVP. Is this correct? Is there really no option to run imMVP() without imputing missing data?

[YinLiLin]

In the function 'MVP.Data()' of rMVP, we have implemented very simple procedure to impute missing genotype, thus it acceptable for rMVP to run GWAS using missing genotype. However, we recommend not doing that if the proportion of missings for markers is very large (e.g., > 10%), it's better to impute by something else (like Beagle as you said) to guarantee stable and reliable results.