How run snpnet without penalizing covariates?

[annilk]

Hi, I wonder how to run snpnet without penalizing covariates?

I can see here that in snpnet(), argument p.factor can be used to control penalization and there it says that by default it is 1 for all variables. Does that mean that by default, all variables, including covariates, are penalized? Which values I should provide there to leave the covariates unpenalized, and penalize only the SNPs?