What we still don't know

Twenty chapters of mechanism, and the honest close is that a lot of this does not yet predict what will happen to you specifically. Here is where the edges are.

Individual response varies more than anyone would like

Give the same fibre, at the same dose, to two people, and the short-chain fatty acid response can differ more than twofold. Not a rounding error — a different result from the same intervention.

The reasons follow from Chapter 14. Fermentation depends on which organisms are present and which enzymes they carry. If someone's community is thin on a keystone degrader, a substrate that depends on it is largely unavailable to them. If the acetate-to-butyrate relay is under-populated, feeding the first step does not produce the second.

Transit time compounds it. The same fibre in a fast gut and a slow gut ferments to different extents in different places.

None of this makes the mechanisms wrong. It makes them population-level statements. "This fibre is butyrate-favouring" is a claim about what tends to happen, not a prediction about your next month.

What the tests cannot yet tell you

The obvious response is to measure your microbiome and personalise from there. That is a reasonable instinct and the products exist, but the science is not where the marketing is.

Sequencing tells you which organisms are present, roughly, in one stool sample, at one moment. It tells you much less about what they are doing, and reproducibility across samples from the same person is imperfect. Interpretation is the harder problem: mapping a species list onto a recommendation requires knowing which configurations produce which outcomes, and that mapping is mostly not established.

Personalised prediction has genuinely worked in one adjacent area — individual glucose responses to meals have been predicted from microbiome and other data with real accuracy (Zeevi et al., 2015). Notice what makes that tractable: the outcome is measurable, immediate, and continuous. "Which fibre is right for me" has none of those properties.

So the honest read is that a test can currently tell you your community differs from someone else's. It cannot reliably tell you what to do about it.

Where the mechanism-to-outcome step is missing

Worth being explicit about which claims in this course rest on what.

Well established. Butyrate is the main fuel of colonocytes. Viscous fibre slows gastric emptying and glucose absorption. Bile acid entrapment lowers LDL cholesterol. Fermentation front-loads and declines along the colon. Cross-feeding produces much of the butyrate. Fibre intakes are far below recommendations almost everywhere.

Good mechanism, evidence mostly outside humans. The oxygen–HIF-1α barrier loop. Mucus thinning under fibre deprivation. Butyrate's effects on gene expression through HDAC inhibition.

Mechanistically sound, not demonstrated as an outcome. That deliberately extending fermentation further along the colon changes how a person feels or fares. This is the load-bearing idea behind designing a fibre for the whole path, and we have been clear throughout that it is a design argument rather than a proven result.

That last category is not a reason to ignore the reasoning. Mechanism is how you decide what is worth testing. It is a reason not to let anyone — us included — describe it as settled.

Go deeper: three open questions worth watching

Does distal delivery matter as much as the mechanism suggests? Targeted-delivery molecules of the kind used in the propionate work make this testable in a way it was not a decade ago.

Can response be predicted before the fact? Detecting whether someone carries the relevant degraders and cross-feeders is a narrower and more tractable question than general microbiome interpretation.

How much does the community actually change, and for how long? Diet shifts composition within days, but much of the shift reverses when the diet does. Which changes persist, and what determines that, is unresolved.

The mechanisms are real and the population-level associations are strong. The step from "this is how it works" to "this is what it will do for you" is the one nobody has finished, and you should be suspicious of anyone who says otherwise.

What to hold on to

You started this course with the idea that fibre is roughage that helps you go. You should be leaving with something considerably more specific: a class of plant structures varying independently along four properties, feeding a community of specialists that hand products to each other, along a gradient with a beginning and a far end, producing molecules your own cells burn and your own hormones respond to.

And with a working sense of where that knowledge stops. That second part is not a weakness of the field. It is what tells you which claims to trust — including ours.