Lecture 4 Flashcards
Substution of a pyrimidine to a purine is called —, and pyrimidine/ purine to a different variant of itself is called —
transversion, transition
what are pyrimidines in DNA?
T and C
What are purines in DNA?
A and G
In the JC69 model with t reaching infinite amounts, how will the nucleotides bee distributed?
it reaches stationary distribution where any long sequence will be composed of equal amount of each nucleotide, no matter what the initial distribution was.
How do we express the likelihood of throwing x times a 6 out of n tries in a binomial distribution as a function?
slide 26 lecture 4
what is the log likelihood of the function above?
slide 26 lecture 4
How do we obtain a confidence interval based on log likelihood?
-we determine the value of the log likelihood function in parameter estimate l(estimate;x)
- we subtract 0.5 X2_k_5% from the found likelihood, we then determine those values for which the condition above holds for the likelihood of the parameter itself.
what are some limitations with the JC69 model in comparison to real life?
in this model all sites in the sequence evolve at the same rate but substitution rates might differ across the genome, as : 1- mutation rates might differ across sites
2- selective pressure might be different on the phenotypic level
How do we overcome the limitation on varying genome rates across sites?
by replacing the constant rated by random variables which are tau distributed with mean on 1
Ignoring site variation leads to ——-?
underestimation of the sequence distance
Are all phylogenetic reconstruction methods based on distances?
No
Can we use the same Markov model used for amino acid substitutions? if yes, are there any adjustments we should make?
yes, the substitution matrix and transitional matrix would both have dimensions of 20 x 20.
What are two types of methods for amino acid substitutions?
empiric, where we try to estimate from sequences rates and mechanistic, where we use a model framework .
For amino acid substitution, do we prefer methods with time reversiblility or not?
yes.
if we have 20 AA, what its the substitution rate for any substitution?
19 lamba
what is the expected time until a substation happens in 20 AA?
1/19lamba
To estimate the distance with an empirical or mechanistic Q matrix we use —?
MLE
We have — possible combination of nucleotides with — amino acids. M eaning —
64, 20. different combinations of nucleotides encode for the same amino acid
In a codon sun, what’s important to note?
where a gene starts and where it ends.
What are synonymous substitutions?
A substitution at which the amino acid doesn’t change e.g TTA and CTT, which still encode for CTA
What are nonsynanomous substitutions?
A substituion at which the amino acid does change.
Why do knowing whether transversion or transition occurs is important?
It is important because it gives an idea of evolutionary trajectories and evolutionary pressure.
What do we incorporate in the basic model of codon substitution?
k: transition/transversion rate ratio, w: nonsynonymous/ synonymous rate ratio, pi: equilibrium frequency of codon I consisting of nucleotides i1,i2,i3, with equilibrium requesting of each, where pi_I = 1/C*multiplicaiton of the equilibrium frequencies of each nucleotide.
synonymous substitutions do/dont? change the protein, thus they are seen as —-?
dont, neutral
nonsynonoumous substitutions do/dont? change the protein and —–, can act on the new protein.
do, selective processes
To discover selection what needs to be compared?
amounts of non-synonymous and synonymous substitutions.
What are the steps for finding the distance at the syn/nonsyn codon positions?
First, we devide the aa to codons with 3 nucleotides and we compare the corresponding condons with their amino acid, if they’ve changed it’s non-syn, if they led to the same amino acids, it’s sin. if there are two or 3 codons different we have to find the average for each. we then find the whole possible number of syn/non syn for each codon and we average it by 3, finally we combine them all for the whole AA as N for average non-syn substitutions possible and S as average syn substitutions possible for N, finally we use the distance formula to find the ratio
if dN/dS<1 ?
non-syn substitutions occur less frequently than syn mutations –? purifying selection
if dN/dS>1 ?
non-syn substitutions occur more frequently than syn mutations –? positive selection ( evolutionary pressure for them to arise)
What are the differences between the JC69 and the TN93 models?
In the JC69 models the substitution rates are the same however, in TN93, the rated vary depending on the type of transition and transvertion, additionally the long term frequency distribution is also parameterised.
Which one of JC69 and TN93 would one choose if they want to perform a phylogenetic analysis based on sequence distances?
TN93, since JC69 lacks basic features of realised, and it doesn’t allow between transition and transversion, and we have to account for the fact that transversions are much harder to occur than transitions.
In reconstructing a phylogeny based on three sequences, would we expect that the nucleotide sequence models would result in different trees?
yes, due to the dependancy on type of differences between a pair of sequences, other than just solely depending on the exact number of differences.
