## Reconstructing Trees An Example

One way to reconstruct phylogenetic trees is to use maximum parsimony analysis. In this method, you determine the minimum amount of evolution required to explain a particular character set. The tree with the minimum number of evolutionary events is called the most parsimonious, or shortest, tree.

You can reconstruct a phylogenetic tree in other ways, such as maximum likelihood, Bayesian analysis, and UPGMA (Unweighted Pair Group Method with Arithmetic mean). Each method has its advantages and disadvantages, but I'm not going to explain them all. I cover parsimony because it involves the smallest amount of mathematics. (Google maximum likelihood, and you'll see that you're getting off easy!)

The following sections take you through a simple example of tree construction.

### Identifying characters

Suppose you want to reconstruct a phylogenetic tree for Species A, B, and C, using seven different character states (1 through 7). For this example, what the characters are doesn't matter; the characters could be eggs present or absent, hair present or absent, and so on.

This example is uncomplicated by homoplasies (refer to the earlier section "Looking at homoplasies" for info), but phylogenetic reconstruction can get very complicated very quickly. Fortunately, for the purposes of this book, you don't need to know how to reconstruct complex trees. I just want you to understand the basic process and to realize that reconstructing these trees isn't magic.

### Assigning polarity

Through comparison with the outgroup species (X), which has the ancestral character state for all seven characters under consideration, you assign a character polarity to each one. The number 1 represents the derived character state, and the number 0 represents the ancestral character state.

Looking at the characters for the other three species, suppose that you find the following:

1 For characters 1 and 2, all three species have the derived condition, making species A, B, and C different from species X (the outgroup).

1 For characters 3 and 4, only species B and C have the derived characteristics.

1 For characters 5, 6, and 7, only species C has the derived characters.

### Grouping species

Because species A, B, and C have the derived character for characters 1 and 2, you know that they are a monophyletic group that doesn't include species X. On the phylogenetic tree, you indicate characters 1 and 2 with two slash marks — one labeled with the number 1 and another labeled with the number 2 — at a point below the common ancestor of A, B, and C (see Figure 9-8). It's most parsimonious to assume that these two evolutionary events happened only one time.

Because only species B and C have the derived character states of characters 3 and 4, you separate these two species as a monophyletic group and indicate the characters 3 and 4 with slash marks below the point of their common ancestor (see Figure 9-9).

In the case of characters 5, 6, and 7, only species C has the derived characters. Because these derived traits aren't shared with any other species, they don't give you any information about the topology of the tree. (Remember, to group species within a tree, you must have shared characters.)

Placing these characters on the tree adds to the total tree length, which is defined as the total number of changes required to explain the data matrix, which in this case is seven (see Figure 9-10).

Characters

Figure 9-8:

Character states for different species.

Taxon

Characters

Taxon

1

2

3

4

5

6

7

0

0

0

0

0

0

0

1

1

0

0

0

0

0

1 1

1 1

1 1

1 1

0 1

0 1

Species A, B, and C comprise one mono-phyletic group; species B and C another.

Characters
 Taxon 1 2 3 4 5 6 7 X (outgroup) 00 0 0 0 0 0 A 11 0 0 0 0 0 B 11 1 1 0 0 0 C 11 1 1 1 1 1

Figure 9-10:

Only species C has derived characters 5, 6, and 7.

 Taxon 1 2 3 4 5 6 7 X (outgroup) 0 0 0 0 0 0 0 A 1 1 0 0 0 0 0 B 11 1 1 0 0 0 C 11 1 1 111

A word about more complicated trees

The preceding example includes only homologous characters — those that appear as a result of sharing a common ancestor. When homoplasies are involved (similar characters that don't indicate a common ancestor), things get more complicated.

If you assume that sharing similar characteristics means that species belong in monophyletic groups and plot the species accordingly, you end up in with a tree that doesn't make sense: A species may appear in different groups, for example, indicating that they evolved more than once.

¿jtJABEft To avoid such a scenario, you need to remember that similar characters can evolve independently, or they can appear, disappear in subsequent evolutionary changes, and then reappear. When multiple scenarios are possible, the one(s) you support end up being those that have the fewest (or most parsimonious) evolutionary steps.

0 0