How To Interpret The Differences In The Phylogenetic Tree Structures Obtained Using Different Software.

Question

[Pappu]

I have constructed phylogenetic tree using NJ method in ClustalX and ML in PhyML. I also mapped my protein sequences belonging to NCBI taxons to the tree of life. Now I am wondering how to interpret the difference in tree structure obtained using these two methods.

[lh3]

Gene trees do not necessarily agree with the species tree due to duplications, losses and incomplete lineage sorting. As to different tree builders, they differ because their models are different and may focus on different information conferred by the data. The general consensus is that ML is better than NJ, though occasionally NJ may produce better trees.

Answer 1

[Manu Prestat]

Comparing a species tree (taxonomic) and a gene tree is something interesting. You would find a lot of informations and paper by looking up "species and gene tree reconciliation" methods. Another important explanation not mentioned by Heng is horizontal transfer of genes, and comparing both trees is a way to detect such an event.