Haplotype-aware trio analysis

[Ricky Magner]

Hi all,

I was hoping to try out using `rtg mendelian` to do some trio analysis, and was wondering to what extent other rtg tools use haplotype-aware comparisons like `rtg vcfeval` does. In particular, if I do `rtg vcfmerge` followed by `rtg mendelian`, is it true that either tool will be careful to merge/compare according to the underlying haplotypes similar to `vcfeval`, or will it just crudely look at the raw GT values and do a comparison? If not, if there some other sequence of tools/flags I can do to ensure this level of detail to the analysis? I did some preliminary comparison and found a number of sites where the raw GTs point to a MCV but the underlying haplotypes are actually consistent.

Best,

Ricky