I am trying to set up a relatively basic model which I can later build upon by adding more parameters to be estimated. I have three .nex files with rRNA, tRNA, and protein-coding gene partitions for 25 species imported into BEAUti, and a relatively simple Birth-Death model wherein
- Clock model (strict) and substitution models (HKY) are unlinked
- Trees are linked across partitions
- Clock rates are to be estimated
Unfortunately, I get the rather strange error of treeLikelihoods being NaN. I saw here some information about resolving this type of error, but I have not specified any MRCA priors, or included any time-dependent information (other than the clock rates, which at init values of 0.01 are adequate for the timescales I am interested in (millions of years).