How do diss.dist() and prevosti.dist() "handle" missing data?

[Mari West]

Hi everyone, 

I am working with a filtered high-density rad-seq ant genomic dataset and am looking into diss.dist() and prevosti.dist() to calculate absolute genetic distance, as an estimate of within-colony pairwise relatedness. I am not interested in population-level relatedness and the genetic population structure of my dataset is quite heterogenous, which is why many other relatedness metrics don't work very well with my dataset. 

I am trying to decide whether I should use the missingno() function to remove all loci with missing data from my dataset before calculating genetic distance. The poppr documentation suggests that diss.dist() and prevosti.dist() can "handle" missing data, so this may not be necessary, depending on what is meant by that. For example, if missing data is ignored (i.e. no comparisons between individuals at loci where the data for one of the individuals is missing), that would be ideal. However, if missing data is "handled" by guessing what the most likely allele (i.e. imputation or something like it), this could introduce error into my analysis, particularly because of the aforementioned heterogeneity of my population sample. Does anyone know what exactly the poppr documention means when it says that these functions can "handle" missing data?

Many thanks,

Mari