There’s too many stories in modern medicine still that sound like, “if we can’t see it, you’re not sick.” There’s at least one more h pylori out there somewhere.

The vast majority of GIs would agree with the grandparent post, but it even goes deeper than that. Pathogenicity of common bacteria is heavily mediated through HGT and interspecies microbial signaling - meaning that there’s plenty we don’t know about even the most studied organisms when they’re in a dynamic environment with other neighbors.

In other words, we know there are multiple layers of activity taking place that we’ve barely scratched the surface on research-wise, so we have good reason to believe that simpler explanations are almost certainly reductive and wrong.

The results have kept coming and it's hard not to believe the gut-health connection now but it's still frustrating that it doesn't translate to any therapy.

Compost does great things for plants, but you can’t sprinkle mycorrhizal fungi or banana peels on top of barren earth and call it a day. Similarly, you can’t throw probiotic bacteria or kimchi into a GI tract and call it a day.

Instead, in both cases, you have to holistically balance and change inputs over time to see results, taking into account the unique, pre-established state of the system. That makes for difficult research when so much of labwork is about controlling and isolating variables, and IMO, is the biggest factor in why we’re not seeing those pre-packaged applications yet.

I've been working on a monologue about reductivism vs reductionism and it reminds me of Doug Hofstader setting up a dialectic between reductionism and holism with the problem that you can't say holism has the same presence that reductionism has.

But soil health really is a place where holism reigns, thanks for reminding me!

https://en.m.wikipedia.org/wiki/Microbial_dark_matter

They’re usually only observable in the rare case that they get randomly captured on an electron microscope slide. The genomic method is complicated because there’s so much data it’s hard to extract the signal and reassemble the snippets. There’s already more data than we know what to do with.

That said, recently a fellow HNer used the genomic method on old data and may have discovered a new class of life one step down from viruses: https://news.ycombinator.com/item?id=42547489

In the human context we are much closer to being able to culture anything we want. I would estimate that we are at 75% coverage now -- pick a species detected genomically in the human gut (or skin/vagina/nose, etc.) and we are likely to be able to culture it or a type strain of its species 75% of the time. For a long time we overestimated human microbial diversity because of bad genomic methods and error rates with the early 454 and then Illumina sequencers. This error rate, coupled with the fact that we can't easily replicate culture conditions for the vast majority of earth's bacterial biomass (i.e. the deep lithosphere or the oceans) led to this persistent if somewhat untrue description of the unculturability of human microbes.

[0] gives an estimate of 35-65% of human/mouse gut microbes having a culturable representative. This paper is from 2017, and there have been a lot of advancements since then.

[1] is a good non-review paper where they got 73% of the genomically defined species via just a single cultivation medium

There are others, but I finished my PhD in culturomics-related stuff in 2021, and haven't kept up as much as I should. Everything the OP cited about lithospheric, deep ocean, etc. still applies as far as I know. Those environments are just very hard to recreate with their pressure, temperature, and nutrient requirements.

[0] https://doi.org/10.1038/s41564-023-01426-7 [1] https://www.nature.com/articles/nature17645

Edit: typos