3 ms·

I guess if it does NOT work on mice, it's highly unlikely to work on humans? In this case, it might be good predictor of failure, which can save a lot of money.

I guess if it does NOT work on mice, it's highly unlikely to work on humans? In this case, it might be good predictor of failure, which can save a lot of money. Just speculating... Also, if something really bad goes on mice, we can avoid having that happening to humans. Which I would also value a lot. I mean, it seems to me experimenting on mice might be worth it.

ekianjo 5y ago
> I guess if it does NOT work on mice, it's highly unlikely to work on humans? There is virtually no relationship whatsoever between what works in mice and humans. Such headlines are completely ridiculous by now: the track record of applicability is a train wreck.
- BizarroLand 5y ago
  No, this is wrong. The point is to figure out what chemicals react with living creatures in what way. We're not creating super-mice, we're not making the rats of NIMH, we're blindly fumbling about for one of the septillions of light switches we haven't flipped yet to see what it does. Figuring these things out is hard. Humans are good at doing things they've done before, or imitating actions they've seen or heard before, but learning something entirely new is very very difficult. We have more scientists alive and working on problems now than ever before in human history and still the mill of innovation turns slowly.
- JamesBarney 5y ago
  This very much depends on the mouse model of disease. Obesity drugs for instance are highly predictive.