Lecture 3 – Molecular phylogenetics

Question 1

what is phylogenetics?

Answer 1

reconstructing patterns of shared ancestry between organisms, either among or within a species

Question 2

what is taxonomy

Answer 2

describing, naming, and classifying species- the organisation of organisms based on phylogenetic/other info

Question 3

orthologous sequence

Answer 3

from different species- used to look at speciation, extinction etc

Question 4

homologous sequences

Answer 4

from the same species- can be used to look at population genetics

Question 5

paralogous sequences

Answer 5

different genes within the same genome- can be used to look at gene duplication, deletion etc

Question 6

advantages of using molecular characters ratjer than morphological ones

Answer 6

-they are more objective and therefore easier to quantify
-available even when morphology is uninformative, e.g. for microorganisms
-cheap and fast
-don’t require specialist training to obtain

Question 7

disadvantage of using molecular characters rather than morphological ones

Answer 7

can’t be used to look at extinct species, most of the time- leaves gaps in phylogenies etc

Question 8

different types of SNP

Answer 8

transition- purine-purine etc
transversion- purine-pyrimidine or vice versa

Question 9

what is sequence alignment

Answer 9

alignment of multiple genetic sequences based on positional homology- the idea that there will be conserved sequences at set positions

Question 10

how computer programs are useful in sequence alignment

Answer 10

there are multiple possible alignments for each 2 sequences, so algorithms are useful to determine the likelihood that each alignment is the correct one

Question 11

things that can complicate alignment

Answer 11

long indels, a lot of genetic diversity

Question 12

examples of programs used for alignment

Answer 12

clustal and muscle

Question 13

what is p-distance

Answer 13

proportion of mismatched sites, very simple measure of genetic difference

Question 14

what is the multiple hits problem

Answer 14

once you get to high observed genetic changes, the actual number of changes is probably higher- points where there have been multiple substitutions can go undetected

Question 15

things that help solve the mhp

Answer 15

generating nucleotide substitution models, so you can project the observed distance onto the likely actual distance

Question 16

tools useful in nucleotide substitution models

Answer 16

Jukes-Cantor model, which looks at appropriate nucleotide substitutions

Question 17

amino acid substitution models- useful things

Answer 17

JTT matrix- looks at the actual frequencies of substitutions, with rates obtained from a large survey of protein variation

Question 18

assumptions within JTT matrix

Answer 18

evolution at each site occurs at the same rate
nucleotide base species are always the same for all species
evolution at each site is independent- can’t really avoid this one, but sometimes it isn’t true, e.g. if there are secondary nucleic acid structures

Question 19

how can among-site variation be accounted for>

Answer 19

gamma distribution model- this can model the heterogeneity in site evolution in a fairly accurate way, helping to create a more accurate level of change- genetic distances tend to be higher using these models

Question 20

what are boostrap values on a phylogenetic tree

Answer 20

measure of phylogenetic uncertainty

Question 21

rooted vs unrooted tree

Answer 21

rooted has an evolutionary direction, and only horizontal lines represent genetic distance
unrooted tree- no direction, and all lines represent genetic distance

Question 22

algorithmic methods- how it works, example

Answer 22

genetic distances for each pair are ‘clustered’- e.g. neighbour-joining

Question 23

optimality methods-how it works, example

Answer 23

score to all possible trees based on data, and an optimisation algorithm finds the highest scores. maximum parsimony, maximum likelihood, bayesian inference

Question 24

statistical methods- how it works, example

Answer 24

probability for each possible tree- more of a formal statistical problem. maximum likelihood, bayesian inference

Question 25

maximum parsimony tree- principle

Answer 25

tree which requires the fewest evolutionary changes is the best one, fast but not good for high divergence

Question 26

maximum likelihood tree- principle

Answer 26

finds the tree which is most likely to have led to finding the observed species using nucleotide substitution models

Question 27

bayesian inference- principle

Answer 27

looks at probability distribution, rather than the probability of individual trees- similar to max likelihood

Question 28

what is a parsinomy score

Answer 28

minimum number of evolutionary changes required to explain observed characters- the scores can be added together on a tree

Question 29

what is a ‘hill climbing’ method?

Answer 29

searches through trees using trial and error, but doesn’t check through all trees- just ones that may get closer to the optimum