Data
The capsaicinoid pathway
Two branches converge on one condensation step, and almost all the control sits on one of them. This is the map every project in the catalog is ultimately arguing about.
2
converging branches
~90%
flux via vanillylamine
PAMT
amine-branch control
pun1
synthase locus
01
The shape of it
Capsaicinoids are made by joining two things the plant builds separately. One branch produces a vanillylamine moiety out of phenylpropanoid metabolism. The other produces a branched-chain fatty acyl group out of amino acid metabolism. They meet at a single condensation step, and what comes out the other side is the molecule family responsible for everything the genus is known for.
The architecture is the reason the pathway is tractable to engineer at all. A convergent pathway with one junction gives you a small number of places where intervention is meaningful — unlike a long linear chain, where control is smeared across every step and nothing moves the output much.
02
The parts that matter
The vanillylamine side
Phenylpropanoid intermediates are routed toward the amine precursor. This branch carries roughly ninety percent of the flux control in the models, which is why nearly every heat-scaling effort ends up here.
PAMT
The aminotransferase step on the amine branch. In host after host, modelling puts the ceiling at or near this reaction, making it the first thing a scaling workflow examines.
The acyl side
Branched-chain amino acid metabolism feeds the fatty acyl moiety. Chain length and saturation here are what distinguish capsaicin from dihydrocapsaicin and the minor capsaicinoids.
Capsaicin synthase, pun1
The condensation step that joins the branches, encoded at the pun1 locus. Loss of function at pun1 is the classic explanation for a sweet pepper: the machinery is present, the junction is not.
Ratio, not just total
The mix of capsaicin, dihydrocapsaicin and the minor species determines how heat arrives and how long it stays. That mix is set on the acyl side, largely independently of the total.
Tissue and receptor
Accumulation is tissue-specific, concentrated in placental tissue, and the sensation runs through TRPV1. Both facts constrain design: a promoter in the wrong tissue produces nothing anyone tastes.
03
Why the asymmetry matters
- Converging branches
- Vanillylamine and acyl
- Junction
- Capsaicin synthase, pun1 locus
- Dominant control
- Vanillylamine branch, ~90% of flux
- Primary lever
- PAMT activity
- Ratio control
- Acyl chain length and saturation
- Accumulation site
- Placental tissue
- Receptor
- TRPV1
Roughly ninety percent of flux control sitting on one branch is the single most useful fact in the dataset. It means the design space is far smaller than the enzyme count suggests, and it explains why effort spread evenly across the pathway produces disappointing results while effort concentrated on the amine branch tends to move the number.
It also explains the division of labour between total heat and burn profile. Pushing the amine branch raises the ceiling; adjusting the acyl side redistributes what arrives. A design that wants more heat and a different character is really two separate interventions, and treating them as one is a common way to get neither.
04
How the platform uses it
The pathway is encoded in every host model, native in the Capsicum species and grafted on in the rest. Flux analysis tests it against the host’s budget, variant scanning concentrates on its control points, guide design targets its loci, and the blend controls work on its output ratio. Prediction is then anchored to HPLC measurement from grown material, which is what keeps the whole structure honest.
Reading the pathway, browsing its enzymes and running flux analysis against it are all free. Credits apply only when a question becomes structural.
Ready to run it
Try this in the Splice Platform.
CPU engines are free to run. GPU engines run on prepaid credits.
Related