Lecture 16 - Patterns and Models

Question 1

What is the approximate time to fixation for neutral alleles in diploid populations?

Answer 1

4N generations.

Question 2

What is a substitution model used for?

Answer 2

Used to describe the patterns and rates of sequence change over evolutionary time.

Question 3

What does Juke’s and Cantor’s one-parameter model say?

Answer 3

All substitutions will occur with equal probability (i.e., no bias in the direction of change).

Only one parameter here - the rate of change of one nucleotide to another (α, alpha).
- Thus 3α is the rate at which one nucleotide can change to any other
- Probability of the nucleotide staying the same after one unit of time = 1 - 3α

Question 4

Is the assumption made by Juke and Cantor (i.e., that all base substitutions are equally likely) a realistic assumption to make?

Answer 4

NO - because like-to-like changes are more probable.
- Transitions
- Transversions

Question 5

Is the assumption made by Juke and Cantor (i.e., that all base substitutions are equally likely) a realistic assumption to make?

Answer 5

NO - because like-to-like changes are more probable.
- Transitions
- Transversions

Question 6

What is a transition?

Answer 6

Like-to-like substitution.
- Purine to purine (A <-> G)
- Pyrimidine to pyrimidine (C <-> T)

Question 7

What is a transversion?

Answer 7

A pyrimidine to purine change (or vice versa).

Question 8

What does Kimura’s two parameter model say?

Answer 8

Assumes that like-to-like changes are more probable.
- This is seen in nature, as transitions are observed TWICE as frequently as transversions
- Thus, mutational bias towards transitions

Question 9

What frequency will each reach nucleotide eventually (in both models)?

Answer 9

25% frequency (i.e., equilibrium).

Question 10

What is the “multiple hits” phenomenon?

Answer 10

The more time that passes, the higher the chance that the same site will undergo multiple substitutions.

Question 11

When is it more likely that multiple hits have occurred?

Answer 11

The higher the divergence, the more substitution events occurred, thus the more likely it is that multiple hits have occurred.

Question 12

What problem arises from multiple hits?

Answer 12

The number of differences that are observed might not necessarily reflect the number of substitutions that have occurred.

Thus, by simply counting the differences, one can extremely underestimate the amount of evolutionary change!
(SEE LECTURE 16 part 1 @ 18:30 mins)

Question 13

How do sophisticated models attempt to take multiple hits into account?

Answer 13

By looking at the degree of divergence between the two sequences.

Question 14

What problem occurs in these sophisticated models (that take multiple hits into account) when looking at high levels of divergence?

Answer 14

Saturation.
(SEE LECTURE 16 part 1 @ 21 mins)

Question 15

When do our substitution models work best?

Answer 15

When the degree of divergence is relatively low.

Question 16

What are the two major violations to the assumptions made in our models (that are seen in nature)?

Answer 16

1. Probability of mutation can vary across the gene or genome regions depending on base pair composition and other factors (e.g., chromatin organisation)!
2. Probability of fixation can vary across the gene or genome regions, due to differing strengths of purifying selection!

Question 17

Give an example of the variability of mutation rates within genomes.

Answer 17

CpG dinucleotides (i.e., C followed by G in sequence).
- Susceptible to methylation!
- Cytosine deaminated into methylcytosine
- Methylcytosine more likely to give rise to T substitutions

Rate of transitions at CpG sites is approx. 18 times higher than non-CpG sites!

Question 18

How does strength of purifying selection vary within the genome?

Answer 18

1. Mutations in protein coding regions are more likely to be deleterious (thus will be removed by purifying selection).
   - Protein coding regions are the slowest evolving regions
2. Even within genes, some regions are under less selective constraint than others!

Question 19

What is a missense mutation?

Answer 19

A nucleotide substitution that result in an amino acid change.

Question 20

What is a nonsense mutation?

Answer 20

A nucleotide substitution that results in a premature “stop” codon.

Question 21

What are the 3 amino acid sequences that encode STOP codons?

Answer 21

1. UAA
2. UAG
3. UGA

Question 22

What is the amino acid sequence that encodes a start codon?

Answer 22

AUG

Question 23

What is a non-synonymous mutation?

Answer 23

A nucleotide change that alters the amino acid sequence.
- Missense
- Nonsense

Question 24

What is a synonymous mutation?

Answer 24

A nucleotide change that does NOT alter the amino acid sequence.

Question 25

What codon position is often synonymous?

Answer 25

3rd codon position.

Question 26

What codon position is NEVER synonymous?

Answer 26

2nd codon position.

Question 27

What codon position is sometimes synonymous?

Answer 27

1st codon position.

Question 28

For synonymous sites, what is rate of substitution?

Answer 28

Rate of substitution = Rate of mutation (µ)

Question 29

What are non-degenerate sites?

Answer 29

Sites at which any nucleotide change will result in an AA change (i.e., never synonymous, 2nd codon).
- LOWEST rate of substitution observed here

Question 30

What are two-fold degenerate sites?

Answer 30

Sites at which one of three possible nucleotide changes results in AA change (i.e., sometimes synonymous, 1st codon).
- INTERMEDIATE rate of substitution observed here

Question 31

What are four-fold degenerate sites?

Answer 31

Sites at which no nucleotide change will change the AA (i.e., often synonymous, 3rd codon)
- HIGHEST rate of substitution observed here

Question 32

What percentage of AA changes are deleterious?

Answer 32

Approx. 80%

Question 33

What percentage of AA changes are neutral?

Answer 33

Approx. 20%

Question 34

What percentage of AA changes are advantageous?

Answer 34

0.0% (rarely do we see advantageous changes!)

Question 35

How do we detect selection?

Answer 35

By comparing Ka to Ks.
- Ka = number of non-synonymous substitutions per non-synonymous site
- Ks = number of synonymous substitutions per synonymous site

Their ratio (i.e., Ka/Ks) is the test for selection!
- Ka/Ks ratio, also known as the dn/ds ratio

Question 36

What would we expect if no selection is occurring (when comparing Ka and Ks)?

Answer 36

Ka/Ks should be roughly equal to 1.

Question 37

If Ka/Ks ratio is much less than 1, what does this indicate?

Answer 37

Ka/Ks «< 1 = suggests negative/purifying selection is occurring!

Question 38

If Ka/Ks ratio is greater than 1, what does this indicate?

Answer 38

Ka/Ks&raquo_space;> 1 = suggests positive selection is occurring!

Question 39

Although Ka/Ks being greater than 1 is rare, why is it particularly interesting?

Answer 39

As it may indicate adaptive evolution!
- E.g., FOXP2., the “language gene” in humans!

Question 40

What are some issues with the Ka/Ks ratio?

Answer 40

SEE LECTURE 16 part 2 @ 26:30 mins!!!

Question 41

Why is it harder to identify homology in non-coding regions?

Answer 41

As they evolve faster!