I worry that once again the nuance of scientists saying "this paper does not definitively show that the D614G mutation causes increased transmission & there are other explanations" is getting lost in the media frenzy around 'more transmissible strain'
2/NN
2/NN
In short, the paper shows that a mutation from a 'D' (aspartic acid) to a 'G' (glycine) in Spike protein position 614 seems to repeatedly become the more prevalent variant of the virus, when both are present in the population. (D (original) in orange, G (new) in blue)
3/NN
3/NN
One explanation for this (the one in the paper) is that the 'G' mutation makes the virus more transmissible & so it out-competes the 'D' viruses, making it 'take over'.
As the mutation is in a protein key to cell binding, this isn't an unreasonable hypothesis.
4/NN
As the mutation is in a protein key to cell binding, this isn't an unreasonable hypothesis.
4/NN
However, as explained well by @trvrb, this D to G mutation took place at the root of what seeded most of the European (& later much of the US) outbreak:
5/NN
5/NN
So an alternative hypothesis is that the patterns we see are essentially explained by epidemiological patterns - early outbreaks tended to be introductions from Asia & be 'D' viruses, later outbreaks were seeded by intros from Europe & were 'G' viruses.
6/NN
6/NN
Looking at cities/states/areas where we have good sampling over long periods & where both strains were circulating can help to tease some of this out - we might then expect to be less biased if both circulated at the same time - but it's not perfect
7/NN
7/NN
We still could be being confounded by differences in intro patterns in these cities - it's still true that the intro timing of the 2 versions of the virus differs, as did containment - we don't have a situation where both were introduced at same time, same conditions.
8/NN
8/NN
The paper also talks about CT values - this is a measure of how much virus is present in a sample. The idea is more virus = more transmissible. (Confusingly, lower CT values = more virus!)
The preprint shows a significantly lower CT (more viral load) for type 'G'
9/NN
The preprint shows a significantly lower CT (more viral load) for type 'G'
9/NN
This is replicated by work from @fredhutch -
@wcassias & @pavitrarc.
However we must also be cautious here - there are confounders in who was hospitalised when, time to collection & how comparable CT values are over time.
10/NN
@wcassias & @pavitrarc.
However we must also be cautious here - there are confounders in who was hospitalised when, time to collection & how comparable CT values are over time.
10/NN
Perhaps more importantly, we don't yet know exactly what the relationship is between CT value & real-life transmissibility. And we don't know if the small but significant difference we see above actually translates into any functional difference in transmission.
11/NN
11/NN
So, this pattern is certainly intriguing. But it far from proves that we have a version of the virus that is 'more transmissible' - mostly the preprint shows patterns, for which there is another, equally valid explanation. We can't tell the two apart at the moment.
12/NN
12/NN
Scientists have been upset by the certainty that the preprint authors have declared that there's a new more transmissible strain, when many believe this is far from as certain as portrayed.
biorxiv.org/content/10.110…
13/NN
biorxiv.org/content/10.110…
13/NN
And it's easy to see why this bothers scientists when there's media coverage like the story in @latimes.
Nuance is lost - people get the idea there's a more 'dangerous' virus, & people get worried & upset. We then spend a *lot* of time trying to put the nuance back in.
14/NN
Nuance is lost - people get the idea there's a more 'dangerous' virus, & people get worried & upset. We then spend a *lot* of time trying to put the nuance back in.
14/NN
These kinds of back-and-forth among scientists happen *all the time.* Someone publishes something that many think is taking the evidence a step to far - we have a lot of discussions about it - we consider what we know & don't know, & we try to work towards an answer.
15/NN
15/NN
It's rarely that they are so public, or that they have such an immediate effect on the public, or that the publication immediately burdens many of us with the next Sisyphean boulder to explain again & again the knowns & unknowns, every day, to a worried public.
16/NN
16/NN
So feelings run hot. But the underlying discourse - the discussion & disagreement, the fact that something could be wrong & not proven *or* right & not proven - this is normal. This is science *working*.
See @edyong209's great piece for more: theatlantic.com/health/archive…
17/NN
See @edyong209's great piece for more: theatlantic.com/health/archive…
17/NN
To end, I want to highlight & recommend a couple more threads which are good reading on this topic of much discussion:
Trevor's thread, which I've already quoted a few times above:
18/NN
Trevor's thread, which I've already quoted a few times above:
18/NN
@angie_rasmussen covers alongside a similar story in Ebola & has a great take on the media coverage, 'correlation != causation' & the difference between being wrong & being unproven - & why scientists need to be cautious how they convey that:
18/NN
18/NN
