4 ms·

I'm hesitant to take an anonymous and unsourced internet comment at face value, but assuming it's true, that is quite damning for American academia! It occurs t

I'm hesitant to take an anonymous and unsourced internet comment at face value, but assuming it's true, that is quite damning for American academia! It occurs to me that the claimed effect, if it exists, might also be due in part to simple expertise. Or rather a lack of expertise among grad students and to a lesser extent post docs. Is anyone here able to corroborate the claim made in OP's linked comment? And does anyone know what kind of experience and oversight the researchers doing discovery in big pharma will have?

Brakenshire 8y ago
I can’t corroborate it except with another anecdote, a medical researcher told me the same thing in passing. She had worked in academia and in industry, and said that industry doesn’t remotely trust any academic work, and that this had been valid in her experience as well. For example that at the university lab that she joined there was a measurement device which had been regularly used, but which no-one understood how to calibrate properly. I suspect as you say it’s partly down to a lack of expertise, being thrown in at the deep end without much guidance on how to get good measurements, both practically and in terms of avoiding statistical traps like cherry-picking, allied to the pressure to find some exciting results, and journals’ lack of interest in negative results.
bhk 8y ago
People I know in big pharma companies have told me they first try to reproduce promising academic results before they invest a lot of money in further research, and that 70%+ of the time they fail to reproduce the results. This came up in conversation years ago when the "reproducibility crisis" first became a hot topic. Other acquaintances of mine, who are in academia, did not find these results surprising at all. In fact, one added the observation that even when the experimental results can be duplicated, the conclusions of the study are probably still not valid because alternative, more mundane explanations are often overlooked, and the peer review process these days doesn't apply sufficient scrutiny or skepticism toward this. I don't have any anecdotes on US vs. Japanese research.
Balgair 8y ago
I'll echo the same from other commentators. There are a LOT of issues with reproducbility in Bio [0] at the moment, and in US based science in general. From others in 'Big Pharma' that I know, they agree that much of bio-tech based science needs to be re-run. Others in the comments section here point out that Mainland China has a history of suspect research. That may have been true 5 years ago, but they are 'up to par' with US based researchers now. Meaning, that the Chinese have gotten more rigorous and the US has severely slipped in quality. I mean this next statement with full honestly: I don't believe half the labs out there in the US these since about 2016 onward. From very private conversations with others, many labs are abusive, fraudulent, 'me-too' landmines, and generally lying to grant funders, donors, and potential/enrolled students in a strange game akin to 'Catch Me If You Can' style con-artistry. Major efforts are being taken to 'fix' the system. Me-Too is a perfect example. The Nature paper that is linked is another perfect example. These efforts are very much needed and those that undertake these efforts should be applauded and emulated. That said, Max Planck's quote is the speed at which this change will occur, namely, one funeral at a time[2]. [0] https://www.reuters.com/article/us-science-cancer-idUSBRE82R12P20120328 https://www.reuters.com/article/us-science-cancer-idUSBRE82R... A bit old, but should get you started on more up-to-date research. [1] https://www.theatlantic.com/health/archive/2018/09/what-is-food-science/571105/ https://www.theatlantic.com/health/archive/2018/09/what-is-f... Food Science, but it illustrates the structure that leads to fraud. [2] https://en.wikiquote.org/wiki/Max_Planck https://en.wikiquote.org/wiki/Max_Planck
- earthicus 8y ago
  > From very private conversations with others, many labs are abusive, fraudulent, 'me-too' landmines, and generally lying to grant funders, donors, and potential/enrolled students I have a friend who worked in a neurology lab a few years ago who described the PI in a similar way. I had assumed that it was not normal and told her to get the hell away from that lab asap... how sad that this kind of shortsighted idiocy is so widespread!

Brakenshire 8y ago

I can’t corroborate it except with another anecdote, a medical researcher told me the same thing in passing. She had worked in academia and in industry, and said that industry doesn’t remotely trust any academic work, and that this had been valid in her experience as well. For example that at the university lab that she joined there was a measurement device which had been regularly used, but which no-one understood how to calibrate properly. I suspect as you say it’s partly down to a lack of expertise, being thrown in at the deep end without much guidance on how to get good measurements, both practically and in terms of avoiding statistical traps like cherry-picking, allied to the pressure to find some exciting results, and journals’ lack of interest in negative results.

bhk 8y ago

People I know in big pharma companies have told me they first try to reproduce promising academic results before they invest a lot of money in further research, and that 70%+ of the time they fail to reproduce the results. This came up in conversation years ago when the "reproducibility crisis" first became a hot topic. Other acquaintances of mine, who are in academia, did not find these results surprising at all. In fact, one added the observation that even when the experimental results can be duplicated, the conclusions of the study are probably still not valid because alternative, more mundane explanations are often overlooked, and the peer review process these days doesn't apply sufficient scrutiny or skepticism toward this. I don't have any anecdotes on US vs. Japanese research.

Balgair 8y ago

I'll echo the same from other commentators. There are a LOT of issues with reproducbility in Bio [0] at the moment, and in US based science in general. From others in 'Big Pharma' that I know, they agree that much of bio-tech based science needs to be re-run. Others in the comments section here point out that Mainland China has a history of suspect research. That may have been true 5 years ago, but they are 'up to par' with US based researchers now. Meaning, that the Chinese have gotten more rigorous and the US has severely slipped in quality. I mean this next statement with full honestly: I don't believe half the labs out there in the US these since about 2016 onward. From very private conversations with others, many labs are abusive, fraudulent, 'me-too' landmines, and generally lying to grant funders, donors, and potential/enrolled students in a strange game akin to 'Catch Me If You Can' style con-artistry. Major efforts are being taken to 'fix' the system. Me-Too is a perfect example. The Nature paper that is linked is another perfect example. These efforts are very much needed and those that undertake these efforts should be applauded and emulated. That said, Max Planck's quote is the speed at which this change will occur, namely, one funeral at a time[2]. [0] https://www.reuters.com/article/us-science-cancer-idUSBRE82R12P20120328 https://www.reuters.com/article/us-science-cancer-idUSBRE82R... A bit old, but should get you started on more up-to-date research. [1] https://www.theatlantic.com/health/archive/2018/09/what-is-food-science/571105/ https://www.theatlantic.com/health/archive/2018/09/what-is-f... Food Science, but it illustrates the structure that leads to fraud. [2] https://en.wikiquote.org/wiki/Max_Planck https://en.wikiquote.org/wiki/Max_Planck

earthicus 8y ago
> From very private conversations with others, many labs are abusive, fraudulent, 'me-too' landmines, and generally lying to grant funders, donors, and potential/enrolled students I have a friend who worked in a neurology lab a few years ago who described the PI in a similar way. I had assumed that it was not normal and told her to get the hell away from that lab asap... how sad that this kind of shortsighted idiocy is so widespread!