Thread by @tony_vandongen

Omicron: Silent Mutations Redux.

Omicron appears to have very few "silent" (S) mutations, compared to the large number of “functional” (NS) amino acid-changing mutations

The ratio dN/dS in Omicron is 41/4 = 10.25

Is there something wrong with that?

If you make random substitutions of nucleotides in the genomes of corona viruses you get a dN/dS ratio of ~3.5

This is to be expected, because changing the 3rd base of a codon usually doesn’t change the amino acid.

That is not that different from 10.25 in Omicron, is it?

For a proper perspective. lets look at the dN/dS ratio for all the corona virus predecessors of Omicron.

The table below shows the "percentage of non-synonymous mutations" in the genomes of Alpha through Iota.

Many zeros, highest is 12%, average is 1.5%

To compare the 1.5% of functional (non-synonymous) mutations in all the predecessors of Omicron, we need to convert its 10.25 ratio to a percentage: 1025%

OK. 1025% is quite different from 1.5%

What happened?

Of course there are several explanations.

One of them has been offered in a previous thread on this topic: Omicron resulted from a chronic infection of an immunocompromised person (possible HIV-1 patient).

We need to talk about that.

The other issue is: why are the NS percentages so low (1.5%) in corona viruses (other than Omicron), when most nucleotide substitutions alter the amino acid they encode?

Answer: selection pressure is different for S and N mutations.

Most amino acid changes do not enhance evolutionary fitness of the virus, or are even deleterious.

Those will not be able to favorably compete with wildtype viruses or mutations that do enhance the virus's ability to multiply. Typical Darwin scenario.

What about the neutral, S mutations?

They almost never cause a deleterious effect.

Unless they produce a codon with a really low "bias", or alter the RNA structure to enhance destruction or inhibit translation.

Both are rare events. So they are mostly not selected against.

The end results of the unbalanced selection pressure against silent (S) and productive (N) mutations, is that the dN/dS ratio tends to get really small, like the 1.5% seen for the Omicron predecessors.

So again: why is Omicrons dN/dS ratio 1025%

In order to come up with the 33 successful mutations observed in Omicron, a great many other mutations would have to be introduced.

You cant have 33 "lucky guesses".

Of course we don't know yet how many of the 33 new mutations in Omicron improved fitness.

We do know by now that Omicron outcompetes Delta.

So some of these mutations certainly improved fitness.

In order to calculate how many silent (S) mutations we expect, given the 33 de novo functional (N) mutations in Omicron, we need to know how many were introduced and subsequently removed.

Because every 3.5 N mutation creates 1 S mutation.

But even in the highly improbable "33 lucky guesses" scenario, where every mutation is either harmless or improves fitness, you would expect 33/3.5 = 9-10 silent mutations. And we only have 4.

So in a more probable scenario, we expect many more than 10 silent mutations. Why does Omicron not have enough silent mutations?

Darwin not at work here?