Biology Reference
In-Depth Information
as fixation frequencies in a population, and by nonlinear statistical effects
of signal saturation when the accumulated number of mutations becomes
too high.
4. Tree Reconstitution
After preparation of the input data, a reliable method is needed in order
to create a tree from the selected characters. This section discusses these
methods.
Many different algorithms exist, and explaining them all in detail is
beyond the scope of this chapter. However, most of the algorithms can
be grouped together into three main categories. For alternative methods,
see Semple and Steel.
1
Before we get down to the details of the three categories, we will
describe the theoretical framework common to them all.
4.1. The Tree Graph Model — Transmission
of Phylogenetic Information
As we have seen, a tree graph, as opposed to a reticular (network-like)
graph, has certain inherent properties. For instance, there can only be a
single path from one node to another. This holds true for all nodes,
whether internal or terminal. Furthermore, when time is taken into
account in a phylogenetic tree, information always flows in the same direc-
tion. Branches are unequivocally oriented from the root to the terminal
nodes (in Fig. 7, from node “A” along the respective lineages to “C”,
“D”, and “E”).
In genealogy (the study of family trees), it is possible for two repre-
sentatives of two separate lineages to breed and thereby reconnect their
lineages. By contrast, in the phylogenetic model, which applies to a
higher taxonomic level than the individual, in principle lineages cannot
cross each other. Horizontal gene transfer
m
is sufficiently rare to be
m
Horizontal gene transfer is the exchange of genetic information through means
other than simple heredity. This usually involves transposable elements, viruses, or
bacteria. The result is a reticulate phylogenetic network that is no longer a tree.
