Thread by @BillHanage

There is a new ‘variant’ clearly identified today, P.1. And it is worth saying a little about what we have learned about it, and from the variants we have identified so far virological.org/t/genomic-char… 1/quite_a_few

First these are not simple ‘mutants’. Mutations happen all the time and most mean nothing much and either persist at low levels or are removed by selection. The variants are characterized by *multiple* mutations – one of the reasons we call them variants and not mutants 2/n

Those multiple mutations are in parts of the genomes that are not exactly the same, but they overlap to a really pretty marked degree (esp the receptor binding domain). This looks like convergent evolution in which different lineages find the same solution by different routes 3/n

(Think bats and birds and their wings for an eg of far more distantly related things coming to a similar solution. One of my rules of evolution is there's usually more than one way to skin a cat, and we can learn from the different ways evolution has skinned cats in the past) 4/n

The variants we are thinking about now all arose independently, but they have things in common. Notably a mutation in the spike protein at the 501 amino acid position. We write it N501Y as shorthand 5/n

Now N501Y is a mutation that has been reported lots of times, because it has happened lots of times (the many millions of cases worldwide do provide the opportunity to mutate) but hasn’t always taken off. Why not if it is allegedly so awesome? (<-rhetorical question) 6/n

It is worth noting that in each of the currently identified variants, this mutation exists alongside *many* others in the same protein and elsewhere in the genome. Almost like it’s not strong enough on its own and needs its friends for support 7/n

This is a comment on a couple things in evolution. First if a mutation happens all the time and is good, why is it not everywhere already? It should have outcompeted everything. One possibility why not is, it has negative impacts too and there is a trade off 8/n

And the fact that mutations at one place in a protein have an impact depending on what else is going on in the protein. They can and do interact. A mutation might be awesome in one context, and meh in another 9/n

It is telling that all the variants reported so far contain *many* mutations over expectations. It suggests quite strongly that the regular rate of evolution is not equal to the task of producing them. That said, they *have* emerged so how did that happen? 10/n

Well they don’t emerge often. There are certainly more than the 3 identified to date, yet to be found. But how many millions of infections have happened in the meantime since the start of the pandemic? 11/n

In some contexts, like this (which I worked on w many smarter people) the clock rate (sort of but not quite the same as the mutation rate) is clearly faster in patients who are chronically infected. nejm.org/doi/full/10.10… 12/n

This is reasonable. For the virus to survive in a host it needs to adapt which drives a higher clock rate. So the ‘variants’ may have spent time evolving in a context like that. Not necessarily *that*one (chronically infected patients) but one w similar selective pressures 13/n

And variants matter. Look at the UK. Not just now but over the next month which I expect to be terrible (please do check back – will be very surprised if I am wrong). B.1.1.7 is rampant now and the consequences will be horrific 14/n

But the crucial message remains that this is a respiratory virus with ways to control it. There are countries that *have* controlled it. If yours has not, ask why not. Those who cannot work from home deserve protection too, and that starts with keeping transmission low 15/end