Hacker News Reader

Comment by andy_ppp | Hacker News Reader

I’m pretty sure if we aren’t careful the same replication crisis that happened in social science will happen here, the interaction of the gut bacteria is so complex it will be very difficult to get consistent results.

View on HN

I wonder how far away we are from simulating the human gut, so we can introduce a variable, hit 'run', and take a snapshot every x hours.

I'm guessing we're 0.0000001% of the way there, given how complex all the interactions must be.

I work in a lab where some of the researchers are working on the much simpler fly gut. Even in the simpler fly It’s very difficult to get anything approaching causality because they’re are so many variables in gut bacteria populations. It’s a huge ecosystem unto itself.

Flies can eat literal garbage so I am also curious how their guts do not have arbitrary digestive problems (let's just say there's not much quality control on the input). Also whether they develop infections.

I looked around a teeny bit and there's some stuff but what you describe doesn't show up right away.

This is a chemical (?) simulator that's been around since 1993 I believe: https://www.wur.nl/en/product/Simulator-of-Human-Intestinal-...

I also found this numeric simulator: https://learn.genetics.utah.edu/content/microbiome/simulator... but I can't view the content right now.

They already do this to design fighter jets, rockets and space suits.

Those are much, much simpler systems. Your average human microbiome has 30+ trillion cells in it, too say nothing of the intestinal cells. That's not getting into the underlying complexity of the cells themselves.

I think we're still far away from that, but we're trying to get there. Some successful work has been done based on AGORA models [1], where the simulations seem to roughly match the reality of simplified systems. For example, for a five-member community of bacteria, answering questions like "what happens if we remove bacteria B from the community?", but not too much beyond that yet.

[1] https://www.nature.com/articles/nbt.3703

Reply on news.ycombinator.com