Hacker News Reader

Comment by hyrix | Hacker News Reader

In mice

Mice are in the same clade as primates, and share around 95-99% of the same genome as primates.

We're all part of the Euarchontoglires clade

The problem is that primates are pretty far removed from mice. And our sheer weight difference often makes is a decisive factor.

This is especially true for cancer, brain, and inflammation research. Mice just don't make good models for that.

> that primates are pretty far removed from mice

They are a good initial test for viability because of the overwhelming genetic similarity.

This is why mice and rodents are used to test initial viability before then moving on to theraputics development.

You can read some sources if you want [0][1][2]

[0] - https://www.jax.org/why-the-mouse/excellent-models

[1] - https://www.cam.ac.uk/research/research-at-cambridge/animal-...

[2] - https://www.genome.gov/10001345/importance-of-mouse-genome

> This is why mice and rodents are used to test initial viability before then moving on to theraputics development.

They are used because they're the _only_ way to test drugs early in development. Animal cruelty laws make it nearly impossible to use non-murine species for initial drug tests.

Additionally, mice are easy to breed and to keep, and there's a great variety of specialized genetic lines.

Absolutely, yet the underlying similarity between our genomes makes it good enough for most cases, plus the Rodent genome has been sequenced and tested for decades now.

Idk why you are trying to die on this hill. I've provided decently authoritative sources (The Jackson Institute, Cambridge University, and the NIH)

Unless you have a PhD in Systems Biology or Genomics, I'm going to trust them over you.

Edit: an actual OG in the field of genomics just commented - listen/chat with him, not me.

For most cases, yes. Antibiotics discovery? Sure, mice are awesome.

But mice make really bad models for anything to do with brain, aging, inflammation, and cancer. That's why comments "in mice" are appropriate in these areas.

And it's not at all controversial, here are my citations:

https://www.pnas.org/doi/full/10.1073/pnas.1222878110 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6615039/

Mice studies are not useless, for sure. But they're almost always are not directly translatable to humans.

As a layman, why should I trust your citation over studies that highlight the advantages of using mice in aging research? - [0]

> But mice make really bad models for anything to do with brain, aging, inflammation, and cancer

If mice is really bad for research in the fields you mentioned, why do scientists around the world continue to use it? Wouldn’t they know it’s a dead end?

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3074346/#:~:tex....

Not sure why you think weight is such a key variable. Might make sense if you had focused in bone mineral density—although mice have been extremely useful to study osteoblast and osteoclast biology.

Mice gave us a very solid understanding of cancer genetics. Cancer was initially thought of as an almost purely environmental disease. Thanks to “in mice” we learned that genetic susceptibility is a major factor.

Transplantation biology and immunology have profited enormously from mouse models—-and more “recently” everything from monoclonal antibodies to CAR-T cells.

And when you say “brain” who could been a dozen different things. I can assure you that mouse models are highly useful and translationally relevant to many neurological, behavioral, developmental, and physiological studies of the mammalian brain.

> Not sure why you think weight is such a key variable.

I worked in cancer research-adjacent area (on computational biology). The problem with mice was that cancers in mice necessarily contain fewer cells than human cancers.

So a lot of treatments can simply clear these cancers completely, without giving them a chance to evolve resistance. Yes, you can try to compensate for that, but it just is not reliable.

In addition, mice are notoriously naturally susceptible to cancer. So their cancers, counter-intuitively, are often less evolved as a result.

> And when you say “brain” who could been a dozen different things.

I basically mean "degenerative brain diseases", very much including aging.

Yet none of that matters because only a tiny fraction of interventions make it from rat models to humans.

Ditto from Phase 1 to Phase 3. Side effects always an issue. One could say only half in jest that humans are not particularly good models for humans. Exceedingly expensive and all kind of environmental and compliance confounders.

At least when using rodent you can control the environmental factors in a randomized controller “clinic” trials. Compliance not so hard to ensure.

Reply on news.ycombinator.com