30x gets you in the ballpark. there are always regions that can't be sequenced well and types of mutations that cant be reconstructed from the data. But from a

30x gets you in the ballpark. there are always regions that can't be sequenced well and types of mutations that cant be reconstructed from the data. But from a clinical utility point of view a good 30x genome gets you up into the 95% range for finding the subset of variation that is clinically interpretable at the moment.