"You don't write such applications out of the blue," says Martin Wikelski, Director at the Max Planck Institute for Behavioural Biology in Radolfzell.
"It really worries me for the future that the experiments conducted were not classified as gain-of-function trials by the researchers," said Karin Moelling, Professor of Virology at the University of Zurich.
A big mistake was, for example, the declaration of solidarity by well-known scientists with their Chinese colleagues, which was published in the renowned scientific journal "Lancet" in February 2020 - and which, as is now known, was initiated by Peter Daszak of the EHA.
"If we as scientists start to stop thinking scientifically and questioning ourselves, but instead blindly protect each other, then we are digging our own grave," Wikelski said.
I may be losing track, but it is at least his third retraction.
There is also on expression of concern for one of his papers.
@thackerpd @KatherineEban @emilyakopp
Here is an important reminder to the Kindergarten epidemiologists who aim to compare themselves to John Snow.
Epidemiology 101:
John Snow never considered his map as proving anything. He relied on fortuitous control groups and cases reviews to establish causality
@mvankerkhove
See for instance this image and extract from a recent paper:
Confirmation of the centrality of the Huanan market among early COVID-19 cases
Reply to Stoyan and Chiu (2024) arxiv.org/pdf/2403.05859…
John Snow was not a colourist of maps, sorry.
I know that popular culture has transformed the Broad Street map into a meme, but that is totally wrong and can only hurt the discipline.
@RichardKock6 @JamieMetzl
1/5 It is difficult to be more mistaken than Robert Garry below, when discussing a supposed essential finding of Worobey et al:
@TheJohnSudworth @MichaelWorobey @hfeldwisch
2/5 As a matter of fact, that pattern is exactly the one expected if proximity to the market was used as a criteria when identifying cases (as is amply recorded).
Going further, there is no easy way to explain that pattern otherwise.
Polling must have been done before Oct 2023, so before:
- Key Science erratum for Pekar et al (invalidated their model)
- Peer reviewed paper showing key statistical flaw in Worobey et al
- DEFUSE draft showing planned work at P2 in China and more
3/26 Then we need cumulative numbers to express the results in a natural way:
- For 19% of experts, a research accident is at least 50% likely
- For 44.6% of experts, a research accident is at least 20% likely
- For 61.3% of experts, a research accident is at least 10% likely
Some key points in relation to DEFUSE and the latest revelations about it:
- The ODNI (Office of the Director of National Intelligence) likely never provided DEFUSE to the intel agencies before their Aug 21 report assessments for the Biden intel report.
- Between Aug 21 (end of agencies drafting) and Oct 21 (when ODNI released its long form Biden report), ODNI did not ask either for an update based on DEFUSE we (DRASTIC) published in Sep 21.
- ODNI did not ask for an update after Oct 21 either, to the point where the DoE submitted a revised assessment of its own and unrequested (before 11 Aug 22).