
Solutions
For hot sauce R&D
Heat is a formulation variable, and most of the industry has no way to specify it precisely. The catalog spans 100 SHU to 16,000,000 SHU with burn profile treated as a separate dimension from intensity.
1,000
constructs
70
series
6
tiers
HPLC
heat rating
01
The problem
Every other ingredient in a formulation has a specification. Heat usually does not. A product development team works with whatever the crop gave them that season, and the year-to-year variation gets absorbed by blending, by extract, or by quietly adjusting the recipe and hoping nobody notices.
The deeper problem is that Scoville rating alone is a thin description. Two peppers at the same rating can land completely differently โ one arrives immediately and disappears, the other builds slowly and stays. That difference is the capsaicinoid ratio, it is designable, and almost nobody specifies it because there has been no vocabulary for doing so.
02
What the platform does about it
Heat as a target
Name a rating anywhere from 100 SHU to 16,000,000 SHU and the catalog returns what exists near it. Six tiers from Pepper to Tier 5 give the range a shared vocabulary.
Burn profile as its own axis
Front-loaded or slow-building, short or long-tailed. Profile comes from the capsaicinoid ratio and is set largely on the acyl branch, independently of total intensity.
Find before you design
With 1,000 constructs across 70 series, a specification often already exists. Searching the catalog is free, and it is the cheapest possible answer to a formulation question.
Design when it does not
If nothing matches, the designer takes a target and tests it against a host's actual flux budget before anyone commits greenhouse space to it.
Measured, not asserted
Catalog heat ratings come from HPLC on grown material, and the interface keeps measured values visually distinct from model predictions at every point.
Why it feels that way
Capsaicinoids act through TRPV1, and the pathway data explains why a ratio change alters the sensation. Useful when a tasting panel reports something the number does not capture.
03
A representative project
- Brief
- Flagship sauce, consistent year to year
- Heat target
- Tier 3 band
- Burn profile
- Slow build, long tail
- Catalog search
- Series filtered by profile
- Nearest matches
- 4, two with HPLC ratings
- Gap
- Profile right, intensity 20% low
- Resolution
- Design revision, SS-0733 lineage
The search found the right character but not quite the right intensity โ a common outcome, and a good one, because it means the hard part is already solved. Profile is the attribute that is difficult to find by accident; total heat is the attribute that scales predictably once the branch controlling it is understood.
From there the work moved to the scaling workflow, where flux modelling confirmed the host could fund the increase and located the constraint on the amine branch. The result is a specification a grower can be held to, rather than a description of whatever arrived.
04
Getting started
Start in the heat ladder explorer with the intensity and profile you actually want and see what the catalog already holds. Browsing and searching are free; so is the flux modelling if you move on to designing something new. Credits only enter the picture for enzyme-level work, which most formulation questions never reach.
Designs leave as a complete specification โ sequences, construct records, predicted rating and profile โ that any cultivar engineering lab can take forward. Scoville Splice built its own catalog through exactly this path.
Ready to run it
Try this in the Splice Platform.
CPU engines are free to run. GPU engines run on prepaid credits.
Related