Incongruence between data partitions indicates that the two partitions compared have had different histories or that one of them violate the assumptions of the phylogenetic method [ 29].
To assess congruence between data partitions three methods were used.
We detected significant conflict between data partitions, especially between morphology and molecules.
For the special finite dimensional case of R n in (83), the search algorithm is an iterative algorithm that alternates between data partition and the optimization of a simpler least squares problem.
Also, using prior knowledge enhances the high consistency between data partitioning and domain knowledge.
Incongruence between the data partitions representing different portions of the plastid genome was assessed via the incongruence length difference (ILD) test [ 110], implemented as the partition homogeneity test in PAUP*4.0d102 [ 111].
Although the molecular data rejected the results of MP analyses of all kinds of data, there were fewer significant conflicts between molecular data partitions and phylogenies derived from the combined data.
However, when independent evidence is lacking and incongruence occurs between individual data partitions, it may be difficult to determine whether particular partitions are better estimates of the species tree than others.
Fourthly, we evaluated concordance between the analyses of the separate data partitions, and between the parsimony and Bayesian analyses.
In order to test for congruence among data partitions, 104 replicates of the partition homogeneity test (ILD) [ 74, 75] were run in PAUP 4.0b10, comparing the phylogenetic signal within each of the following partitions: a) between both mitochondrial genes, b) between both nuclear genes, c) between mitochondrial versus nuclear genes, and d) among all genes.
