Total Evidence, Average Consensus and Matrix Representation with Parsimony: What a Difference Distances Make
Claudine Levasseur and François-Joseph Lapointe
Département de sciences biologiques, Université de Montréal,QC, Canada.
Abstract: Matrix representation with parsimony (MRP) can be used to combine trees in the supertree or the consensus settings. However, despite its popularity, it is still unclear whether MRP is really a consensus method or whether it behaves more like the total evidence approach. Previous simulations have shown that it approximates total evidence trees, whereas other studies have depicted similarities with average consensus trees. In this paper, we assess the hypothesis that MRP is equally related to both approaches. We conducted a simulation study to evaluate the accuracy of total evidence with that or various consensus methods, including MRP. Our results show that the total evidence trees are not significantly more accurate than average consensus trees that accounts for branch lengths, but that both perform better than MRP trees in the consensus setting. The accuracy rate of all methods was similarly affected by the number of taxa, the number of partitions, and the heterogeneity of the data.
Readers of this also read:
- Identification of Conflicting Selective Effects on Highly Expressed Genes
- Environmental Quality, Developmental Plasticity and the Thrifty Phenotype: A Review of Evolutionary Models
- On the Adaptive Design Rules of Biochemical Networks in Evolution
- The identification of functional motifs in temporal gene expression analysis
- Fast Genes and Slow Clades: Comparative Rates of Molecular Evolution in Mammals
