This block type is used to set root lengths and to choose which models are used with which trees. In addition it is used when simulating multi-partitioned datasets where different sections of the simulated dataset are created with different trees and models.

Example Usage:

  [PARTITIONS] partitionname [treename name rootlength]

    [t1  m1  rootlength1]
    [t2  m2  rootlength2] 
    [t3  b3  rootlength3]
    [t4  m4  rootlength4]
    [t5  b5  rootlength5]
  • partitionname is used to identify a particular [PARTITIONS] block in the [EVOLVE] block.
  • treename is the name of any previously defined [TREE] block.
  • name is the name of any previously defined [MODEL] or [BRANCHES] block.
  • rootlength is the length of the root sequence that will be generated for this partition.

  • If using a [BRANCHES] block instead of a [MODEL] block then remember that the trees from the [TREE] and the [BRANCHES] block must match.
  • Because of this it is NOT permitted to use random trees with [BRANCHES] blocks.

  • pM is an example of how to define the [PARTITIONS] block when you want to generate multi-partitioned datasets.
  • The tree, root length and model/branch class can be different in each and every partition and there is no limit on the number of partitions other than that imposed by computer memory.
  • However the trees must all have the same number of taxa and branch-classes can not be used in the same partition as random trees.