Topic 15: Phylogenies

Question 1

What are phylogenetic systematics, phylogeny, and phylogenetic trees?

Answer 1

Phylogenetic systematics: study of evolutionary relationships between organisms
Phylogeny: evolutionary history of a group of organisms
Phylogenetic tree: used interchangeably with phylogeny, a branching diagram depicting the ancestor-descendent relationships among a group of organisms

Question 2

What is a node? What is bifurcating or multifurcating?

Answer 2

A point at which a branch splits into two or more branches, they represent a hypothetical or real ancestor.
Bifurcating is when it splits into two and multifurcating is when it splits into more than 2 (polytomy)
Tips represent extant (living) taxa or OTU (operational taxonomic units)

Question 3

What is a branch and a branching pattern?

Answer 3

Branch: a line depicting the ancestor-descendent relationship between two nodes
Branching pattern: topology or cladogram
Branch length can be used to represent number of changes that occured in that branch.

Question 4

WHat is the difference between a rooted and an unrooted tree?

Answer 4

Rooted: a tree in which the direction of evolution through time is implied (required knowledge of ancestral state)
Unrooted trees: trees which indicate relationships among taxa, but with no directionality

Question 5

What is rooting an unrooted tree like?

Answer 5

It is essentially picking a branch and pulling up by some point.

Question 6

TRUE or FALSE
There are always more unrooted than rooted topologies for any given number of taxa

Answer 6

FALSE
There are more rooted than unrooted

Question 7

To properly root a tree, what is necessary?

Answer 7

We need to know the common ancestor of the group of interest, and this requires an outgroup
Outgroup: taxon or group of taxa that are closely related to our group of interest but known not to belong to the group

Question 8

What are characters?

Answer 8

Anything that can be assessed in the taxa
DNA sequences, morphological traits, behavioural traits

Question 9

What are the three character classifications?

Answer 9

Invariant: character that is the same state in all taxa, NOT USEFUL
Uninformative: character that is variable in state but does not confer any phylogenetic or grouping information, one state may be shared by many taxa but no other state is present in more than one taxa
Informative: a character that has a minimum of two states where each state is shared by at least two taxa

Question 10

What are cladistics? What are phenetics?

Answer 10

Cladistics: trees constructed on the basis of shared evolved characters (Max parsimony, max likelihood, Bayesian)

Phenetics: trees constructed on the basis of similarity, distance-based methods (UPGMA, and neighbour joining)

Question 11

What is a clade?

Answer 11

The set of taxa/OTU derived from a common ancestor, includes all taxa descendent from a particular node but no other taxa (monophyletic)
Extant taxa in a clade,always form a monophyletic group

Question 12

What is maximum parsimony? What are the three steps to finding the most parsimonious tree?

Answer 12

All things being equal, the simplest explanation is the best

1. Construct every possible (unrooted) tree
2. For each possible tree, count the # of changes required for each character and sum over all characters
3. Select the best tree by choosing the tree with the fewest changes

Question 13

How do you find the tree score on a maximum parsimony tree?

Answer 13

Character mapping (set theory )

Question 14

What is set theory?

Answer 14

A set is a collection of elements
S= {A,C,G} is a set of three elements
The null set (circle with line through it) is the set containing no elements
The intersection between sets (upside down U) is the elements contained in both sets
The union (U) of both sets is all the elements in both sets, with doubles counted once
First move from tips to root and label with either the intersection or the union of them.

Question 15

What is an apomorphy/ synapomorphy/ plesiomorphy/ symplesiomorphy?

Answer 15

Apomorphy: derived character state (different from ancestor)
Synapomorphy: apomorphy shared between two or more taxa
Plesiomorphy: ancestral character state (usually inferred from the group)
Symplesiomorphy: a shared plesiomorphy

Question 16

What happens if there is more than one maximum parsimony tree?

Answer 16

They can be combined into a consensus tree. This is a way to illustrate the similarities between the MP trees

Question 17

How does homoplasy affect a tree? What is it?

Answer 17

Homoplasy is when two or more taxa share a character state not through shared ancestry, but through independent events
May cause issues with character mapping

Question 18

What are the three mechanisms of homoplasy?

Answer 18

Parallel evolution: both separately evolve the same state
Evolutionary reversal
Convergence

Question 19

What is maximum likelihood?

Answer 19

Estimates the likelihood of all the character data, given the proposed tree
Likelihood is maximized over all possible topologies/branch lengths/ internal node assignments
Best tree is that in which the data has the highest likelihood of occuring, and branch lengths indicate the number of mutations that have occured on each branch

Question 20

What is a Bayesian method?

Answer 20

Find the most probable tree based on the likelihood and the incorporation of prior information
Best tree is the most likely, given the data

Question 21

What is Phenetics?

Answer 21

The study of relationships among a group of organisms, on the basis of the degree of similarity which may be molecular, phenotypic or anatomical.
Distinct from using character states in cladistic methods
Generally use genetic distances as the information for building trees

Question 22

What is UPGMA? What are the steps?

Answer 22

Unweighted pair-group method with arithmetic mean, given a matrix of genetic distances between taxa

1. Identify from among all OTU’s, which two are most similar (smallest distance apart
2. Create a new node by joining the two most similar OTU’s with equal branch length to each terminal node
3. Treat the two grouped OTU’s as a single OTU and calculate a new distance matrix
4. Repeat steps 1-3 with the new genetic matrix

Question 23

What are the two types of UPGMA ties? How do you deal with them?

Answer 23

When two or more pairs of taxa are the same genetic distance apart.
1. If one species is included in more than one of the shortest clades START TWO TREES
2. If both of the shortest clades contain completely different taxa sets CREATE MULTIPLE UNATTACHED CLADES FOR THE SAME TREE

Question 24

What is entailed in neighbour joining?

Answer 24

Start with a star phylogeny, and minimize the total distance when each pair is pulled out.
This pair that is then joined together to form a new node and the procedure is repeated

Question 25

What are the two methods to assess the statistical significance of a tree?

Answer 25

Consensus (complete and majority rule) as well as bootstrapping

Question 26

What is involved in consensus trees?

Answer 26

Often there are more than one equally good trees, and a consensus tree combines a number of different possible trees to provide a summary of the information
Objective of making this tree is usually to identify groups that are monophyletic in all of the possible trees

Question 27

What is monophyletic, paraphyletic and polyphyletic?

Answer 27

Monophyletic: a group that contains the most recent common ancestor and ALL descendents
Paraphyletic: group that contains most recent common ancestor and NOT ALL DESCENDENTS
Polyphyletic: group that does NOT contain the most recent common ancestor

Question 28

What is the difference between a strict consensus tree and a majority rule consensus tree?

Answer 28

STRICT: only the taxa that are monophyletic in ALL trees are grouped together
MAJORITY RULE: only the taxa that are grouped together greater than 50% of the time are grouped together

Question 29

What is involved in bootstrapping?

Answer 29

Re-sample original data set with replacement to construct a series of replicates of the same size as the original data set (characters)
Each replicate analyzed like the OG
Variation among replicates indicates the confidence we have in the result from the original dataset
Each replicate data set is subjected to parsimony analysis and one most parsimonious tree is maintained
Finally, majority rule consensus tree is constructed for all replicate trees, if a group appears in X percent of bootstrap trees, then the confidence we have with that group is X