Title: Comparative methods: Using trees to study evolution
1Comparative methods Using trees to study
evolution
2Some uses for phylogenies
- Character evolution
- Ancestral states
- Trends and biases
- Correlations among characters
- Molecular evolution
- Evidence of selection
- Key innovations
- Diversification rate
3Why reconstruct character evolution?
4How do we know that bat and bird wings are not
homologous?
5Why reconstruct character evolution?
- Can evaluate homology
- Can determine character-state polarity
6(No Transcript)
7Why reconstruct character evolution?
- Can evaluate homology
- Can determine character-state polarity
- Can evaluate the selective regime when a
character evolved
8Map on selective regime
9Now look at character evolution
10Now look at character evolution
11Why reconstruct character evolution?
- Can evaluate homology
- Can determine character-state polarity
- Can evaluate the selective regime when a
character evolved - Can recreate ancestral genes/proteins
12Character optimization using parsimony
- Pick the reconstruction that minimizes the cost
- What do you do if more than one most-parsimonious
reconstruction - ACCTRAN/DELTRAN
- Consider all
- What character-state weights should you use?
13(No Transcript)
14Cost-change graph(Ree and Donoghue 1998 Syst.
Biol. 47582-588)
15Stability to gainloss weights
16What gainloss weight to use?
- If you believe gains are more common (hence
weighted less) you will find more gains (and vice
versa) - So how can you use a tree to establish if there
is a gainloss bias?
17A likelihood approach
- Developed (in parallel) by Mark Pagel and Brent
Milligan in 1994 - Continuous time Markov model
- Select the rate of gains (0-gt1) and rate of
losses (1-gt0) that maximizes the likelihood of
the data given a sample tree (and branch lengths)
18Transition rate matrix
To
From
19Logic
- Calculate the likelihood of the data for a given
value of q1 and q2 - Modify q1 and q2 to find a pair of values that
maximizes the probability of the data
20Probabilities summed across all possible
ancestral states
1
1
0
1
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
21Are gain and loss rates different?
- Likelihood ratio test
- Model 1 gains and losses free to vary
independently - Model 2 gains and losses equal
- How many degrees of freedom?
- .but invalid due to tree like structure
22Montecarlo approach
- Estimate parameters under model 2
- Use those parameters to simulate evolution up the
tree - Take the simulated data and calculate the LR of
model 1 vs. model 2 - If lt5 of the simulations have a LR greater or
equal to that observed then Model 1 is favored
23Method can be adapted to determine the
probability of an ancestral state
- For example, look at the contribution to the
overall likelihood contributed by state 0 or
state 1 at a node of interest - Can be used as a statistical test of homology
24The likelihood method
- Provides a method for using the data to evaluate
gainloss bias - Takes account of branch lengths
- Sensitive to taxon sampling
251
1
0
1
0
0
0
1
1
0
Suggests that the rate of losses is low
261
1
0
1
0
0
0
1
1
0
Suggests that the rate of gains is low
27Correlated evolution
- Look at pairs of traits (where one trait can be
an environment) - Body size and range size
- Warning coloration and gregariousness
- Fleshy fruit and dioecy
- Do these traits evolve non-independently?
28Non-phylogenetic (tip) method
- Count species
- Do a chi-square test
- But non-independence is a huge issue