Sporadic Distribution Of A Gene On A Phylogenetic Tree

Question

[sanchezcavani]

I am confused by a phylogenetic analysis. My gene phylogeny is shown as the following figure (assume all nodes are well supported). Most species in firmicutes form a clade. However, the gene also shows sporadic distribution in a few distantly related genus including E coli, Thermus Aquaticus, Bartonella, Methanococcu (Archaea), seemingly indicating a result of LGT (lateral gene transfer. However, none of them group with firmicutes, which contradicts the hypothesis of LGT. Could it be because of the accelerated evolution after LGT?

Does anyone have any ideas on how I should explain this phenomenon? LGT, massive loss or others?

[Joseph Hughes]

How did you build your tree: neighbour joining, maximum likelihood, parsimony? Do you have a species tree of the taxa the way they should be related? Can you show a tree with branch lengths?

Answer 1

[Joseph Hughes]

It could be a result of long-branch attraction (LBA), an artefact caused by the rapid evolution of two supposedly unrelated organisms. LBA is discussed in depth in the phylogenetic literature and is not so much of a problem when using maximum likelihood methods that use models that correct for multiple substitutions at the same site.

[sanchezcavani]

Thanks a lot for your comments! The gene phylogeny is quite different from species tree. Several distanly related species are grouped together and form an independent clade (not group together with the species where the gene is mostly found). Basically the branch length of the genes from distantly relating species are very long. So I was wondering if it is due to LBA. However, accelerated evolution has to be assumed to have happened for many times if LBA is used to account for the observed phylogeny.

[sanchezcavani]

Sorry , forgetting to mention that ML was applied to build the tree so that the effect of LBA should be minimized.