Our ensemble doesn't predict Rhinolophus shameli but our two best models - Trait-1 and Network-1 - both do, bringing their hit rate up to 22/24 and 15/15 respectively. (Network-1 still undefeated!) Updates will follow shortly on viralemergence.org/betacov
As the authors point out, SARS-like viruses are still fairly understudied deeper in southeast Asia, but our model predicts that should be a hotspot of undiscovered bat βCoVs....
Increasingly, we think those models might help point to where we should be looking for the progenitors of SARS-CoV-2 in wildlife. And, we think approaches like ours might help tailor sampling (and cut the costs of research funded by taxpayers) in situations like this!
At the same time - this is the first new virus published from a Rhinolophus species (which we assume will probably harbor the SARS-CoV-2 ancestor) since we started tracking new predictions almost a year ago. Tons more work to do on that genus still if we want to find the source.
Getting there requires:
🦇 new fieldwork
🦇 revisiting museum specimens
🦇 getting unpublished data in the open
• • •
Missing some Tweet in this thread? You can try to
force a refresh
A few years back, McNeil wrote a book called Zika: the Emerging Epidemic, which I read and reviewed for Quarterly Review of Biology in April 2018.
I was absolutely shocked with how he talked about race, and about women - both in the abstract, and in his specific interactions.
A couple spicy tweets got some conversation going in my Twitter circles, but that was it. Since then, whenever folks talk about his various odd takes, I've always thrown this example out - and folks are often surprised to see it. (I don't think the book was widely read 🙃)
So post-workshop, the World Meteorological Organization's Task Team on COVID-19 and Climatic, Meteorological, and Environmental Factors has published some guidance on how to do the science. It's nice, but missing the words "talk to an epidemiologist"
It's tough. I appreciate what they've done here, and they very clearly nod to our piece on how climate-but-not-epidemiology experts got things wrong. But I also still think, 10 months in, the magic words are "talk to an epidemiologist about your understanding of the system."
This is particularly salient given that they actively encourage scientists to do public facing communication that "...includes informing media outlets or policy makers of dissenting views and encouraging the presence of multiple voices in coverage of their work."
I think it's easy to assume more of this is riding on "access and cooperation" between China and the WHO / other countries than history would suggest is actually true (thread)
Take SARS-CoV as a counterfactual, where tracing back to wildlife trade was efficient and transparent. Civets are linked to SARS-CoV before the outbreak ends, and horseshoe bats are implicated as the reservoirs of SARS-like viruses by 2005. Access and cooperation at work! But...
The actual reservoir species isn't fully tracked down and published until 2017. That has less to do with early outbreak transparency, and more to do with the arduous nature of tracing viral origins in the wild:
This week I wrapped up COVID-19 related policy work. Just to quickly pin it, here's a reference thread of my writing about various pandemic topics.
Epidemic forecasts are important, but often fail to translate to on-the-ground decision making. We list a handful of high-priority questions, from basic epidemiology to healthcare data science, that policymakers have been asking us to help them answer.
The wildlife trade is implicated in a tiny fraction of emerging disease outbreaks worldwide (and has no concrete link to SARS-CoV-2). Centering wildlife trade regulation as "pandemic preparedness" undermines the work of global health experts.
The 2014 outbreak in West Africa, the largest to date, was traced back to human-bat contact without any link to wildlife trade. Many scientists find the evidence for this incomplete, but it's probably impossible to know now. (2/4)
Some (not all) outbreaks since 1976 were linked to human consumption of wildlife, especially wild primates or bats as a primary food source. This is not the big, international wildlife trade with a criminal underworld side that conservationists mean (3/4)
New COVID-19 comment: Species distribution models are a great tool, but wrong for a respiratory virus. Here's an explainer of where ecologists went wrong, and why we have to stop right now, before people get killed.
Thanks to @joechip90@BlasBenito@richardjtelford@BobOHara and Nature Ecology & Evolution for working around the clock to get this out there (and to the journal for making it open access).
Bottom line:
Using SDMs for a system like COVID-19 is malpractice, deeply indicative of a lack of understanding about microbiology.
Pushing studies out anyway against the advice of public health experts, at great risk to the public, is careerist misconduct.