Running CAFE5 on large tree

[ToriEggers]

Hi,

I'm trying to run CAFE5 on a tree of ~80 species of nematodes. The runs error after failing to predict lambda after 300 iterations with inf values, even with families of the highest size differentials removed. When I subset the tree to smaller clades of about 12 species, CAFE5 does complete. Do you have any advice? Is this just because it is too large of a tree or too deep in time?

Thanks!