How is missing data hadnled by SNAPP and SNAPPER

[Loukia Spilani]

Hi BEAST users. I am trying to build a species tree and date it based on this tutorial: https://github.com/mmatschiner/tutorials/tree/master/divergence_time_estimation_with_snp_data#q2 

I am working with a ddRAD dataset, so my individuals have varying levels of missingness. On top of that, my different species are not made up of the same number of individuals. Because of these two circumstances, I would assume that there are many (if not all) SNPs that might not be present for all species, especially in those cases were my species are made up of few individuals. After reading this post: https://www.beast2.org/2014/07/21/snapp-handling-missing-data-and-path-sampling-made-easier.html though, I thought that I should still be able to use this dataset, even though it's patchy, becase in it says "In v1.1.6, no sites with missing data are deleted any more, even when this leaves some species having zero lineages. SNAPP assumes the gene tree simply has no taxa in that species any more." I know that is refers to SNAPP but given that SNAPPER is based on SNAPP, I assumed that this would be the case for both.

However, when trying to run the dating tutorial for snapper, all my SNPs were removed (WARNING: removed 1000 patterns (1000 sites) because they have one or more branches without data.). So I am wondering, has the missingness handling become more stringent since that version (I currently have snapp v.1.6.1 and snapper v.1.1.5)? Could it be the case that the problem is not the missingness per se, but the fact that my SNPs might not be informative? 

Thanks in advance!