Thread by @ewanbirney

A brief explainer thread on B.1.1.529, the latest SARS-CoV-2 variant which is throwing up concern after an excellent live streamed press conference from South Africa. TL;DR this variant both has many mutations but most importantly looks like it outcompeting delta in South Africa

Context: I am a expert in human genetics and computational biology; I know experts in viral genomics, infectious epidemiology, clinical trials and immunology. I have some conflicts of interest: I am longstanding consultant to Oxford Nanopore and was on the Ox/Az clincal trail

Background. The SARS-CoV-2 virus is made from RNA (its instruction set) wrapped in proteins. The RNA+proteins of the virus hijack our cells to make more of its RNA and proteins into a virus. This hijacking (infection) causes a response from our immune system.

Sometimes the response to SARS-CoV-2 goes haywire leading to more (and deeper) virus infection and this horrible disease, COVID, which I think of as a severe virus-triggered auto-immune disease.

The RNA is a long chemical polymer made from 4 possible units; as such we give them 4 letters (A,U,G and C; U is the RNA form of T in DNA); SARS-CoV-2 RNA is about 30,000 letters long. About every 3 infection cycles on average one of these letters change due to chance events.

Most of the time these changes (variants) don't change anything about the biology of the virus, but when we read the RNA (also known as "sequencing its genome") we can use these spelling changes to track the evolutionary tree of the virus.

We have collectively sequnced millions of SARS-CoV-2 genomes, and shared them (via the covid19dataportal.org and GISAID), building an ever richer family tree of the virus. We can for example work out therefore where particular instances of viruses came from recently.

We could (and do!) name each sequence, but this gets overwhelming; a neat system (called Pangolin from @AineToole, @arambaut and colleagues) keeps track of the major "families" given systematic if tongue-twisting names honouring the tree, like B.1.1.7 and B.1.617.2

There are alot of these and a whole corner of the internet dedicated to keeping this all straight worldwide (its not easy)

Occassionally the variants change the biology of the virus, broadly in two possible ways - they can change how fast the virus transmits between people (or animals) or can change to what extent the virus looks like previous versions to our immune system. ("immune escape")

When the virus changes biology is usually changes the dynamics of the epidemic - faster transmiting means more control is needed to prevent widespread transmission. Immune escape means the virus may well reinfect people and vaccines are less effective.

At this point, becuase we start using the variant in public discourse, WHO senisbly gives them a designation, being a greek letter - hence Alpha, Beta, Delta etc variant.

Armed with this background, what was announced today? First off - huge huge credit to the South Africa surveillance ( @Tuliodna) and epidemiology for their data, analysis and transparency - none of this is easy, in particular at pace with serious consequences of analysis.

A family of the virus, first identified in Botswana and named by Pangolin as B.1.1.529 was tracked in South Africa. This family has a lot of changes (more than 30) including some changes which have lead to more transmission and some to immune escape

However, it was the SA epidemiology which is important. One change knocks out some of the PCR test sites on the genome (PCR tests are done with belt and braces of 3 usually sites, so one knockout doesn't invalidate the test); this means much of the SA testing scheme can track it.

The SA scientists also did genome sequencing, and they saw very high concordance of the full B.1.1.529 genomes and the S-gene drop out - and we've been here before with the Alpha variant (B.1.1.7, also called the Kent variant) first discovered in the UK,

Using now this proxy they can chart the (likely) growth of this variant across SA. And it is explosive in the context of Delta, the current dominant virus across the world. Here is @Tuliodna

https://twitter.com/Tuliodna/status/1463911577544011780?s=20

This has all the hallmarks of a variant of serious concern. First off, South Africa needs help - even more than a week ago. We need to understand it better, in particular consequences infection of this variant wrt vaccination. In the meantime we need to limit its spread.

In practice this means higher levels of quarantine of people from this region, and in country surveillance using all tools at each country's disposal.

I will leave the last tweet to the excellent @Tuliodna with a plea for help - South Africa did the right thing here (as did the UK with Alpha) of surveillance, prompt analysis and transparency. Huge credit to the SA team.

https://twitter.com/Tuliodna/status/1463911554538160130?s=20