There is a lot of information in the branch lengths, both on gene trees (genealogies) and species trees (phylogenies) – don’t get me wrong – but by topology I only mean the grouping.

In the example in the post, the branch lengths tells you a lot about the ancestral effective population size, but only if you also see different topologies can you be in any confusion about the phylogenetic relationship.

Varying branch lengths alone cannot question which species are more closely related to others. Varying topologies can.

Short branch lengths in the phylogeny, combined with large effective population sizes, is likely to result in different topologies.

Large branch lengths in the phylogeny, even with large effective population sizes, will still give you gene topologies that matches the species topology.

The genealogies – gene trees including branch lengths – will never match the species tree (the phylogeny) regardless of the phylogeny branch lengths.

I hope that makes my terminology clearer. If not, let me know.

Also, let me know if I am being inconsistent with the terminology. I am not as careful with it on the blog as I would normally be :)

One of the most compelling confirmations of the theory of evolution, in particular common descent, was the finding of close concordance between phylogenies derived on the basis of organismal biology and those inferred from molecular data. A wide array …

I hadn’t thought about the interference between tree inference and selection, to be honest. I know that recombination messes up the tree inference, and that changing topologies can look a lot like selection if you test for selection afterwards, but I am surprised that selection itself can mess up the tree inference.

I’ll read the paper and see what happens. Thanks again!

I did some simulations of this a while back, and the inferred genealogy can be really far from any of the true genealogies in the segment. That were simulations with lots of recombinations, though, so how serious it is for the cases they consider, I wouldn’t know.

I plan to look into it, though, when I get the time… which won’t be any time soon, unfortunately, since I am pretty swamped in other projects right now.

I’d sure like to see you get back to it. I know of one published comparison of tree inference methods on something close to molecular data using Avida, an evolutionary simulator where one knows the actual genealogy of the opcode programs that are evolved. It’s here, unfortunately behind a pay wall. The abstract:

Phylogenetic trees group organisms by their ancestral relationships. There are a number of distinct algorithms used to reconstruct these trees from molecular sequence data, but different methods sometimes give conflicting results. Since there are few precisely known phylogenies, simulations are typically used to test the quality of reconstruction algorithms. These simulations randomly evolve strings of symbols to produce a tree, and then the algorithms are run with the tree leaves as inputs. Here we use Avida to test two widely used reconstruction methods, which gives us the chance to observe the effect of natural selection on tree reconstruction. We find that if the organisms undergo natural selection between branch points, the methods will be successful even on very large time scales. However, these algorithms often falter when selection is absent.