I'm not a biologist, but a friend of mine is. From what he told me, it is a pretty common "practice" to alter results in many ways to meet the high standard of scientific reviews, because of the pressure to get a publication.

Photoshopping curves to pimp your results, or simply "omitting" part of your experiments that would make your statistics go down.

Note that most of them is done in moderately good faith : very often, something obviously went wrong in one experiment and doesn't invalidate the general idea behind a paper (i'm not saying it's ethical or even true).

Another thing for people not familiar with research paper : just having your name as a coauthor doesn't mean you have anything to do with the actual research beeing done. Most of the work is done by phds or post-doctoral students.

Lab bosses probably don't have the time to double-check every single result of every single paper, especially in a big labs, but they still appear as co-author.

All of that to say : i wonder what is really specific to Mr Raoult, and what is just the ugly side of a field brought to light because of media focus.

I worked at a very successful research lab in the UC system for a few years. Nearly everything we did was fabricated. We would literally write proposals in a day or two and DARPA would throw 250k at us. We'd have some undergrad f around with a robot and video camera, write a paper and call it a day.

Everything was a business decision. I remember watching a PHD candidate being told he needed to stay another year because basically, we needed him to get grant money. I was an undergrad at the time and was offered a PHD track, but I declined because I didn't want that to be me.

The experience has left me with a deep distrust of academic science.

do you know how prevalent this is ? maybe you met other people in other fields.

I would wager science attracts immoral people at more or less the same rate any other field. Look no futher than the reproducibility crisis.

it's been a few years since the scientific world has shown some cracking, publication number games, elsevier, p-hacking, lack of reproducibility. an ear has ran it's course, things have to be reworked

More details from the tweet author: https://forbetterscience.com/2020/03/26/chloroquine-genius-d...

So there are chances that it's yet another fallacious system induced 'start' that may explode. Also the parallels with current PUSA psyche are too ironic.

The title is editorialized, which goes against the HN rules. Linking to tweets is challenging, as there’s sometimes not a lot of context, but even the tweet doesn’t go as far as the HN title...

Literally the 2nd tweet:

As reminder: Raoult's paper on alleged #chloroquine cure of #COVID19 had rigged, omitted or even falsified data and was published without peer review in a journal he controls. Trial was not randomized, controlled or blinded in any way.

https://twitter.com/schneiderleonid/status/12429058031888220...

Sorry, I did not have a chance to flesh this out. The current title ('questions data in old paper') is much more reasonable to describe a set of PubPeer comments.

I'm not making any claims about the chloroquine study.

The tweet thread/sequence/whatever it's called absolutely does go that far and talks about him fabricating data and figures many times, among many other things.

OP here: the chain of tweets goes much, much further than the the title I used, and I certainly wouldn't call it "editorialized" or breaking HN rules.

When submitting articles, I almost always use the original title, verbatim; but as you point out, that's seldom possible with tweets - TBH, I think the title I submitted was pretty reasonable.

It makes me wonder what their motivation was then, because this whole thing only increased their chances of being exposed.

A lot of times people who do this are self-deluded. They "know" they are right and just fudge it a bit to get that across.

Vulnerable world, meet modern academics.