Slack / Private Message Drop — page 882

[2021-05-27 11:11:35] [Andrew Rambaut] In other news we are tracking a B.1.617.2 variant variant with K417N (Karen) - probably at quite high prevalence in Nepal with lots of travel exports. Looks nasty. [2021-05-27 11:15:41] [Kristian Andersen] Interesting - a couple of B.1.429s with Karen in them too. I thought maybe L452R would be non-conducive to Karen.... https://outbreak.info/compare-lineages?gene=S&threshold=75 [2021-05-27 11:52:03] [Andrew Rambaut] But add 'more transmissible than B.1.17' and it could be a problem [2021-05-27 12:41:58] [Kristian Andersen] The important addition to that question that I think is often forgotten - "in what setting". I can believe that 617 is more transmissible than 117 in certain settings (high natural immunity, high one-shot immunity, mostly), but in a naive population, is it more fit? Not so sure about that. [2021-05-27 12:45:09] [Kristian Andersen] And it's worth pointing out that this is strictly an academic question, not a public health one - it's clear 617 is a problem and we need to act, but that was clear weeks ago. [2021-05-27 13:06:20] [Andrew Rambaut] Just discussing that. My personal hunch has been that B117 is adapted to strong NPI by being more transmissible early but for shorter (causing more damage and thus severity). Make the most of limited number of contacts. 617.2 may have a longer pre-symptom period making it better in a more free-for-all environment (i.e., unfortunate timing as UK restrictions have lifted). [2021-05-27 13:10:29] [Kristian Andersen] Yeah, that's an interesting theory - although it seems like 117 may actually also infect for longer? Did Oli ever do any of the modeling analyses to try and flesh these differences out a little more? [2021-05-27 13:34:01] [Eddie Holmes] I wonder if they have enough material to shotgun sequence the environmental samples from the market? Might tell them want animals were being sold my looking at the expressed host genes (e.g. COII). Doesn't even need to be the virus positive samples although they would be better. [2021-05-27 13:38:16] [Kristian Andersen] Well, fuck me - so obvious and honestly I had not thought about it. Yes, they should and yes that should be totally doable. [2021-05-27 13:39:28] [Andrew Rambaut] Could do a COI amplification - i.e., species barcoding? [2021-05-27 13:47:52] [Eddie Holmes] I'll see if I can get George and Weifeng to do it