Chat message

Slack / Private Message Drop, p.135 [SLACK_000349] · slack_pm:msg:01376

Date: 2020-08-14 14:27
Type: chat message · slack
recipient: Kristian G. Andersen, Robert F. Garry, Andrew Rambaut
speaker: Edward C. Holmes

Recipients on this medium are inferred from channel membership, not per-message addressing.

I'm going to ask around about this. At face value it's either a massive fuck-up by the people analysing the data or it'sirresponsible subsampling from NextStrain.

In context

A single line often inverts meaning once you see what it answers, so neighbouring messages are always shown.

2020-08-14 14:22 Edward C. Holmes open

The story here is that the outbreak in Victoria was due to an escape from hotel quarantine from return travellers.There's even an inquiry going on that the moment. I can see from the tree that you are suggesting that the virus wasin Australia prior to the current outbreak? That is certainly not the narrative here. I will need to ask about this. Toclarify, I'm not involved in any of the Australian analysis: it's being done by the public health people in each state. They've shut me out completely.
2020-08-14 14:25 Edward C. Holmes open

Are we sure that the NextStrain tree is just massively subsampled, missing key background lineages?
2020-08-14 14:25 Kristian G. Andersen open

Yeah, that's what I thought too - returning travelers and quarantine issues. I should have clarified that the statementI shared is from the most recent Nextstrain narrative: https://nextstrain.org/narratives/ncov/sit-rep/2020-08-14.I'm not sure we can really conclude it's local acquired from this data, but I haven't looked at it directly (so I'm not sure whether I agree or disagree with the Nextstrain folks here...)
2020-08-14 14:26 Kristian G. Andersen open

And yes - my concern exactly relates to whether subsampling has taken out necessary context - combined with bias,of course.
2020-08-14 14:27 Edward C. Holmes

I'm going to ask around about this. At face value it's either a massive fuck-up by the people analysing the data or it'sirresponsible subsampling from NextStrain.
2020-08-14 14:27 Kristian G. Andersen open

Whether correct or not, pretty interesting though - especially given the pretty strong political line that this is due toreturn travelers (could be - but I'm also not sure about that...).
2020-08-14 14:29 Edward C. Holmes open

True! I'm on it. Thanks for pointing this out!!
2020-08-14 14:32 Kristian G. Andersen open

You bet - definitely worth an extra (careful...) look. They made some conclusions on our Baja / SoCal sequences that are definitely not correct though, so something tokeep in mind (in this particular point, it's impossible to say whether the virus moved South (as suggested), North, orfrom some completely different location). [shared file(s): Screen Shot 2020-08-14 at 2.29.40 PM.png]
2020-08-14 14:32 Edward C. Holmes open

My money is on NextStrain screwing it up.

Full conversation →