Tsdate

Bugfixes

• Variational gamma uses a rescaling approach which helps considerably if e.g. population sizes vary over time

• Variational gamma does not use mutational area of branches, but average path length, which reduces bias in tree sequences containing polytomies

Breaking changes

• The default method has been changed to variational_gamma.

• Variational gamma uses an improper (flat) prior, and therefore no longer needs population_size specifying.

• The standalone preprocess_ts function also applies the split_disjoint_nodes method, which creates extra nodes but improves dating accuracy.

• Json metadata for mean time and variance in the mutation and node tables is now saved with a suitable schema. This means json.loads() is no longer needed to read it.

• The mutation_rate and population_size parameters are now keyword-only, and therefore these parameter names need to be explicitly typed out.

• The ignore-oldest option has been removed from the command-line interface, as it is no longer very helpful with new tsinfer output, which has the root node split. The option is still accessible from the Python API.