An empirical assessment of the bootstrap support as an indicator of robustness in phylogenetic trees / So Wei Huo
Bootstrapping is a de-facto standard for displaying the validity of a phylogenetic tree. However this may not be the case as bootstrapping assume that the random sequence exist within the genome. In order to examine whether bootstrapping is a valid process, it must be done empirically. Across multi...
| 第一著者: | |
|---|---|
| フォーマット: | 学位論文 |
| 出版事項: |
2012
|
| 主題: |
| 要約: | Bootstrapping is a de-facto standard for displaying the validity of a phylogenetic tree. However this may not be the case as bootstrapping assume that the random sequence
exist within the genome. In order to examine whether bootstrapping is a valid process, it must be done empirically. Across multiple taxa there are minimal fully curated genomes but there are mitochondrion genomes available to test this. A widely accepted evolutionary tree is chosen with 6 taxa - Pan paniscus (bonobo), Homo sapiens (human), Gorilla gorilla (gorilla), Pongo pygmaeus (orangutan), Hylobates lar (common gibbon) and Gallus gallus (red junglefowl) as the outgroup. A comparison of topology and bootstrap support value is executed between the phylogenetic tree of the whole mitochondrial genome with trees built from all the genes in mitochondrial genome. The
result shows that bootstrap support value tend to inflate and larger than the empirical estimate. This suggests that bootstrap support in phylogenetic trees must be interpreted
cautiously and not casually accepted at face value. |
|---|