Phylogenetics Midterm

Question 1

Character

Answer 1

any feature of an organism that can be defined by its states

Question 2

Homoplasy

Answer 2

A character that transitions into and out of a given state on a phylogeny (due to being under strong selection, etc.)

Question 3

Synapomorphy

Answer 3

A shared, derived character

Question 4

Plesiomorphy

Answer 4

The ancestral character state for a particular clade

Question 5

Apomorphy

Answer 5

A specialized of derived character state

Question 6

Autapomorphy

Answer 6

A derived character state that is restricted to one taxon in a particular data set

Question 7

Polyphyletic group

Answer 7

A type of non-monophyletic group in which the group does not include the most recent common ancestor of all members of that group

Question 8

Paraphyletic

Answer 8

A type of non-monophyletic group that contains the most recent common ancestor byt not all descendants of that ancestor

Question 9

Differences between phylogram, chronograph, cladogram

Answer 9

phylogram - proportional branch lengths.
chrono - scaled with time
clado - just relatedness

Question 10

2 main things make a phylogenetic tree

Answer 10

Direction and Acyclic (unidirectional)

Question 11

What is MAXIMUM PARSIMONY

Answer 11

Minimize the number of character state transitions needed to explain the data.
Tree with the smallest number of changes (fewest steps, shortest tree) is favored

Question 12

What’s an examples of parsimony INFORMATIVE

Answer 12

Autapomorphies. Nothing else has them

Question 13

Explain the CONSISTENCY INDEX (CI)

Answer 13

Number of character states -1 divided by number of state changes on the tree

Question 14

Be able to write in Newick format and then draw tree from it

Answer 14

n/a

Question 15

What is it called when all tips are extant?

Answer 15

Ultrametric

Question 16

Issues with parsimony

Answer 16

1. The probability of novel mutations on particularly long branches may draw them together (analogous)
2. When they are long enough, adding more data will only push toward the wrong answer more often (balance)

Question 17

Steps of a Heuristic search

Answer 17

1. Stepwise addition. make a tree
2. swap branches, in various way and at various points.
3. repeat multiple starting points
4. shortest tree is winner

Question 18

What are “islands” in this context?

Answer 18

local optimas

Question 19

How to avoid local optimas

Answer 19

continued swapping, explore other tree space, additions

Question 20

Indel

Answer 20

insertions or deletions in the genetic code of an individual(s)

Question 21

Jukes and Cantor Nucleotide Substitution Model

Answer 21

All substitutions occur at the same rate,

Question 22

Define MAXIMUM LIKELIHOOD

Answer 22

Finding a tree that has the highest probability of giving rise to the observed data

Question 23

Steps to get MAXIMUM LIKELIHOOD

Answer 23

1. make a tree
2. Determine the likelihood of each site
3. Likelihoods of each site in the matrix are multiplied
4. The program optimizes parameters and branch lengths by repeating prior steps to find the where the likelihood is maximized
5. Tree space is searched to find the maximum likelihood tree

Question 24

What’s bootstrapping?

Answer 24

Randomly resampling characters / columns with replacement to get a population of the same size

Question 25

Non-parametic bootstrapping

Answer 25

1. Get your best tree
2. Randomize reweigh characters
3. Repeat 10,000 times
4. Make a new best tree
5. Get majority rule/strict consensus trees

Question 26

Steps in Bayesian Markov chain monte carlo (MCMC)

Answer 26

1. Set Priors
2. Propose a tree and set of parameters
3. Calculate the likelihood of the tree
4. Calculate the acceptance ratio (incorporating Hastings ratio)
5. If ratio is 1.0, accept the move
6. If ratio is < 1.0, accept the move with a probability equal to the acceptance ratio
   Run chain until stationarity is reached
7. Begin recording sampled values with a given frequency (eg every 1000 steps) to build up an estimate of the posterior
8. AT THE END - summarize the post-burnin trees and parameters, this set represents your posterior distribution.

Question 27

Results of MCMC, heuristic search, and boostrapping

Answer 27

MCMC- collection of trees sampled from the posterior probability distribution
Huer - best tree and associated parameter estimates
Boot - Estimated confidence in each node

Question 28

Fossils are often treated as the ______ stem age

Answer 28

minimum

Question 29

Penalized likelihood

Answer 29

basically cuts computing time down.

Question 30

3 claims must be made about stated fossils

Answer 30

1. age
2. diagnosis
3. stem or crown calibrated

Question 31

3 types of molecular clocks. list and define.

Answer 31

Strict- fewer parameters bc only the internal branches are free to vary in length, but not the tips
Correlated relaxed - Rates vary continuously across the tree, and are inherited from their ancestral branches
Uncorrelated relaxed - Rates vary continuously across the tree, but there is no autocorrelation of rates

Question 32

What is meant by “lineage birth” and “lineage death”?

Answer 32

birth - speciation event
death - extinction

Question 33

What are ghost lineages?

Answer 33

Nodes that are “hidden” by extinction
Evidence morphologically or even molecularly, but no physical evidence

Question 34

Diversification ________ through time

Answer 34

decreases

Question 35

2 approaches to understanding rate evolution

Answer 35

1. BAMM -> rates are generally constant, and only rarely jump to new, possibly quite different, values
2. ClaDS -> Rates change frequently, but the magnitude of change from ancestral values is constrained by a probability distribution

Question 36

Draw an apopmorphy

Answer 36

na

Question 37

Draw a homoplasy

Answer 37

na

Question 37

Draw an autapomorphy

Answer 37

na

Question 38

Draw a plesiomorphic triait

Answer 38

na

Question 39

draw a monphyletic group

Answer 39

na