CONCEPTS
In genetics, dN/dS also known as or Ka/Ks, is used to estimate the balance between nonsynonymous substitutions and synonymous substitutions at a particular site in the same period. The ratio is used to estimate non-neutral changes relative to neutral changes and the degree of the selective pressure of a protein-coding gene.
dN(Ka):In a given period time, the number of nonsynonymous substitutions per non-synonymous site.
Advantageous mutations -- positive selection; Deleterious mutations -- purify selection; Change the protein, but not particularly change anything about fitness -- Neutral selections, bebop within the population.
Because the nonsynonymous substitutions are thought to accumulate neutrally. So, dN can be used to scale the different mutation rate of different genes.
dS(Ks): In the same period, the number of synonymous substitutions per synonymous site. Neutral mutations -- Neutral selections.
Nonsynonymous substitutions are neutral mutations, and synonymous substitutions are deleterious or beneficial mutations that could be selected during evolution. So the dN/dS ratio indicates the balance between deleterious and beneficial mutations.
METHODS
Pair-wise or multiple sequence alignment of homologous genes:
Maximum-likelihood methods; Approximate methods; Counting methods
RESULTS
dN/dS : >1, Positive or Darwinian selection; <1, Purifying or stabilizing selection; =1, Neutral selection, but not totally neutral
(The average result of different period time and different regions within the gene may mask the magnitude of the selection.)
And performing a statistical analysis of the results is important.
COMPLICATIONS
Sometimes, the swap frequencies between various nucleotides is different. Because it is rather common for transitions to be favoured over transversions, models must account for the possibility of non-homogeneous rates of exchange.
The mutation frequencies in different codons of the same gene are different.
As time progresses, it's possible for a site to undergo multiple modifications. There is no way to detecting multiple substitutions at a single site, thus the estimate of the number of the number of substitutions is always an underestimate.
LIMITATIONS
The non-coding regions, such as regulatory regions, can't be estimated.
It is difficult to interpret when dN/dS = 1.
The effects of nonsynonymous substitutions are not determined.
Because some deleterious mutations may be weeded out during a number of generations natural selection, it is difficult to use dN/dS in the closely related populations. And the effects of time must be incorporated into an analysis, if the lineages being compared are closely related.
------------------------------------------------------------------------------------------------------------------
CALCULATE dN and dS
FIRSTLY, CALCULATE THE NUMBER OF SYNONYMOUS SITES AND NONSYNONYMOUS SITES
There are 12 sites in the two sequences.
In the first site of codon ACT, it may mutate to CCT, TCT, or GCT. ACT is a threonine, CCT is a proline, TCT is a serine, and GCT is a alanine. If this site has any change from A, the amino acid will change. So the first site A is classified as a nonsynonymous substitution site.
All second positions of codons are nonsynonymous sites. Any second position site change will change the amino acid.
In third site of codon ACT, if the codon start with AC and the third site mutates to any nucleotide, and will not change the amino acid. So the third site of codon ACT is a synonymous site.
Using this method, we can classify 12 sites of this sequence into 4(site number: 3, 6, 9, 12) synonymous sites and 8 nonsynonymous sites.
But in some cases, the result of the nucleotide mutation is uncertain. For example, in the third site of codon TTT, and TTT is a phenylalanine, if the third site mutates, TTC will still be a phenylalanine, TTG and TTA will be a leucine. In this case, the probability of synonymous mutations is one third, and the probability of non-synonymous mutations is two thirds.
SECONDLY, CALCULATE dN/dS
dN = nonsynonymous changes/ nonsynonymous sites
dS = synonymous changes/ synonymous sites
In this case, there are one nonsynonymous changes(11C11T, Pro
Leu) and two synonymous changes(3T
3G and 9G
9C). Because the mutation direction is unknown, use "
" to indicate the mutation process. But mutation direction is not essential for calculating dN/dS.
dN = 1/8 = 0.125
dS = 2/4 = 0.5
dN/dS = 0.125/0.5 = 0.25
THIRDLY, THE MEANING OF THE RESULTS
My own views:
Since synonymous mutations are not selected, dS reflects the frequency of mutations at synonymous sites and nonsynonymous sites. So dS indicate the expected mutation frequency at the protein-coding gene region.
Nonsynonymous mutations are positively selected or negatively selected, and the mutations are weeded out or conserved, so dN indicate the frequency of mutations at nonsynonymous sites that can be conserved during selection.
So we can use dN/dS ratio to estimate the selective pressure. And we have to consider the effects of the number of two kinds of mutation sites, evolution time, the size of the target genomic region, the speed of selection and mutation, and so on.
dN/dS indicates which is faster, selection or mutation.
----------------------------------------------------------------------------------------------------
From the Mohamed Noor's course:
So what's happening here is you're basically looking at an average of an evolutionary process when you're looking at the single dN dS value. So you can tell that there's been a lot of constraint or a lot of rapid evolution, or you just can't tell. dN dS can end up being a little too conservative, especially if you're looking for those adaptive amino acid changes, that high dN dS value. You'll have way too many false negatives. McDonald Craigman test can overcome this problem.
REFERENCES
【1】https://en.wikipedia.org/wiki/Ka/Ks_ratio
【2】https://www.coursera.org/lecture/genetics-evolution/searching-for-natural-selection-dn-ds-s-6VocF
