-
Add a new function to the old release, which allows users to partition their input tree into different subtrees. LSD will estimate a separate rate for each subtree which optimises the objective function over the whole tree.
-
Update README and the help page.
-
Use min(1000,sequence_length) instead of real sequence length to generate simulated trees when computing confidence intervals