Early development · Beta. RaRaMa is in active development. Every result here is a research-use-only prediction, a pre-bench screen, not a laboratory assay or a food-safety, allergen or regulatory determination. Confirm anything you will act on with accredited testing.
Deterministic · structure-only · food & agriculture analytics

Screen it before you make it.

A sequence in: taste, aroma, structure, stability, functionality, allergen and protein QC out. Before a gram is purified.

Every prediction is a run you don’t make. One dossier triages a lab-week of bench work, before the sample is in the machine.

Zeta potential vs pHα-lactalbumin · from sequence
Predicted surface potential versus pH for alpha-lactalbumin, computed from sequence. The curve crosses zero at the isoelectric point near pH 4.6, with the stable-dispersion and flocculation-risk bands marked.
Computed from the sequence. The curve crosses zero at the predicted isoelectric point, the pH to avoid for a stable dispersion. One of nineteen virtual instruments returned per protein.
Why this, why now

Taste is the number one wall to plant-based adoption. We screen the attempt before it is spent.

$32–41B
Flavor & fragrance today. The taste-modulation slice alone runs $7.6B toward $13.9B.
Data Bridge · MarketsandMarkets
$85–175k
And 6 to 12 months for a single new-product-development attempt at the bench.
Industry estimate
~44%
CAGR for precision fermentation, about $4–7.6B today toward ~$114B by 2034.
Fortune BI · Research and Markets
$1.1B
Raised into alternative protein in 2024, fermentation the single biggest slice.
GFI
What the work costs today

Every candidate that reaches the bench is a synthesis, an assay, and a share of a trained panel’s week.

Published 2026 rates for the work a pre-bench screen front-runs, laid out in full, not cherry-picked. The point is not any single line. It is that most of this is spent on candidates that do not work out.

One candidate, carried to a real answerCostTime
Custom synthesis of a single candidate, milligram scale$2,000–15,000~3–4 weeks
Instrumental taste read on an electronic tongue, one sample; the required reference control is billed as a second sample$395, about $790 with the control12–15 business days
One beany off-note marker measured in the headspace$213 per sample1–3 weeks
Consumer panel, 75 to 100 people$6,000–10,000weeks
Trained descriptive panel, the human ground truth: about ten people on payroll, each with 100 to 150 hours of training before they score a single sampleno per-sample price, a standing costweeks to months
Full flavor workup that identifies and rebuilds an aroma from its parts$40,000–120,0006–12 months

Rates: Medallion Labs 2026 catalog (instrumental taste and off-note), LSU AgCenter (consumer panel), and published core-facility and method schedules. The trained panel has no per-sample price anywhere in the world, because it is a team you employ, not a service you buy. Its scarce resource is its weeks, not its invoice.

At program scale

One taste-discovery program spent about $425M over nineteen years screening its way to eight ingredients. A structure-only screen front-runs that same decision in seconds, before a gram is made.

500,000→8
A public, audited program screened over half a million compounds to commercialize eight ingredients.
SEC filings
~$425M
Of research over about nineteen years to do it, with roughly a hundred R&D staff at its peak.
SEC filings
$30M + $32M
A single buyer paid up front, plus committed research funding, for one taste. The willingness to pay is real and nine figures.
SEC filings
seconds
The platform reads taste, off-note, structure and stability from the sequence or molecule alone, on a laptop, at almost no cost per candidate.
The shift

A screen does not replace any of this. It decides what to put in front of it: advance the few candidates worth a bench and point your panel’s weeks and your instrument queue at them, instead of the fifty you are guessing between. The ones that will not work die before the synthesis, the panel session and the regulatory clock are ever spent. The program above is Senomyx, on the public market until its 2018 acquisition; every figure is drawn from its SEC filings.

Three landscapes

The path to market runs differently on each continent, and a screen runs the same before any of them start.

United StatesThe largest market. A new food ingredient can clear the common safety route in roughly six to nine months at no filing fee, but a formal food-additive petition runs up to $10M and four years.
EuropeThe slowest gate. A new flavoring has no self-affirmation route and takes about eighteen months at a minimum before it can be used.
CanadaThe feedstock and the cluster. The world’s largest pulse exporter, home to the plant-protein supercluster this lane is built on, and where the platform is based.

Sources: US FDA and published filings; EU flavoring authorization (EUR-Lex); Protein Industries Canada. A screen is jurisdiction-neutral: it runs the same on a laptop before any of these clocks start.

Reformulation, before the bench

Change one ingredient and you re-match two things at once: how it tastes, and how it behaves.

Cut the sugar and a bitter tail appears. Replace the egg and the gel, the foam and the mouthfeel move with it. Today that is a twelve to twenty-four month loop of re-matching taste and function at the bench, most of it spent on versions that never work.

The screen reads both from the structure, before a batch is made, so you carry the few candidates worth confirming to the bench instead of the fifty you are guessing between. The bench confirms, it does not hunt.

Sugar reduction which sweetener, and the bitter tail to design around.
Animal to plant the plant protein that gels and holds water like the one it replaces.
Off-notes the beany or metallic note named from the molecule, not found weeks later in the headspace.
Allergen swap the family caught from the sequence before a batch ships to a separate buyer.
Who uses it, and why

The teams that pay per candidate, in weeks.

The same read a multi-instrument bench produces, but before anything is synthesized or shipped. In this lane most work is two companies passing physical samples back and forth. We let both sides agree the short-list pre-bench.

Precision fermentation

Read a brewed protein from the gene, before the tank.

For strain and ingredient teams, and their regulatory lead

Mass, fold, unfolding temperature and allergen family from the sequence, so a candidate that would fail at scale-up, or an allergen hit, is caught before a batch is run or shipped to a separate buyer.

~$4–7.6B today → ~$114B by 2034
Plant-based meat & dairy

Texture, before the rheometer queue.

For food technologists and ingredient applications teams

Gelation, water-holding and emulsification from structure. Texture is the number one reason a plant-based product fails on repeat purchase, so this is where launches live or die.

~$30–40B combined market
Pulses & plant protein

The beany off-note, named from structure.

For pea, soy and faba protein suppliers

The green and beany volatiles (hexanal and its family) and the bitter saponins, flagged before a GC-olfactometry bench spends weeks on the physical headspace.

Pea protein isolate ~$2.68B → ~$6.74B
Sweeteners, flavor & masking

Which sweetener, and the bitter tail to design around.

For flavorists and sugar-reduction R&D

Sweetness versus sugar and the off-taste flagged from structure, so trained panels (which take about two months to train and drift with fatigue) run only on the survivors.

Sweeteners ~$114B · F&F R&D ~CHF 500M+/yr

Photos: Pexels and Unsplash (free for commercial use). Self-hosted with credits on deploy.

Successes and failures, before the wet lab

A whey protein’s whole identity, from 162 letters.

Beta-lactoglobulin, the main protein in whey, has been characterized in the published literature for decades: its exact mass, its size in solution, the temperature it unfolds, its fold. A multi-instrument bench’s worth of characterization, all of it reproduced here from the amino-acid sequence alone, in minutes.

RaRaMa · from the sequence
Exact weight18,277 Da
UV concentration constant17,210
Size in solution2.17 nm
Unfolds at73.7 °C
Fold / allergenlipocalin
Measured, in the lab
Mass spectrometry18,277 Da
Published constant17,210
X-ray scatter2.16 nm
Calorimetry~75 °C
Crystal structurelipocalin (Bos d 5)
Vallejo-Cordoba 2008 (mass) · Pace 1995 (constant) · Gottschalk 2003 (size) · Wada 2006 (unfolding) · Brownlow 1997 (fold).

Mass and constant exact, size within 0.5%, unfolding within about 1°C, fold and allergen family correct. All from the sequence.

And the failures, flagged from structure.

The drivers behind famous reformulations, each called from the molecule alone, verified live on the platform.

Banned additive · safety flag

FD&C Red 3

Reactive, genotoxic structural alert. The FDA revoked it in 2025. The hazard reads straight from the iodinated structure, before a ban forces the change.

Banned additive · safety flag

Brominated vegetable oil

Reactive, genotoxic structural alert. Revoked in 2024. The organobromine reads as a hazard from the structure alone.

Off-note · flagged

Stevia (rebaudioside A)

Sweet, with a bitter side-taste flagged. The exact reason sugar-reduction programs moved to a cleaner stevia molecule.

Off-note · flagged

Beany aldehyde (hexanal)

Green and beany. The off-note every pea and soy reformulation has to mask before launch.

A miss, left in

Trans fat (elaidic acid)

We do not flag it, and that is honest. Trans fat is a nutrition and policy ban, not a structural hazard, so a fatty acid reads clean. We catch structural hazards and off-notes, not nutritional policy.

The founder

I observe more than I take part in the fluff of social life. I think differently, and I tend to reach conclusions by atypical paths.

I spent five years maintaining railroad track. Hard physical work, long days, most of them away from home. It taught me to stay on a problem until it’s solved.

Both of my parents had cancer. My mother recovered from lung cancer. My father died during Covid, from several cancers at once. That’s why I take on hard problems.

After his passing, I would reminisce about the times my dad and I read the paper together. One day it was about the Google Chromecast, and he said something about how my generation has everything at its fingertips. I took that to heart and taught myself to build things on a 3D printer, sublimation mugs and lamps, to bring in money for my family.

One was a spherical lithophane, and I wanted color in it. The software said it would take about two thousand layer changes, which looked impossible. So I changed the orientation and split the print in two, and the work dropped to a fraction. Then I looked at the default settings the 3D-printing community had all landed on, and my brain read that number as a physical length, not a setting. That length matched how molecules behave in water. It stopped me cold. I knew it was real before I could prove it, and in my gut it wouldn’t let me go. I spent a long time trying to tear it apart. It held.

I aimed the method at drug screening first, but the barriers there overextended what I could do alone, so I adapted it to food and agriculture. That’s RaRaMa: give it a protein sequence, a SMILES string, or an InChI, and it predicts how that molecule behaves.

Dustin Hansley, founder of Hans-Made Research
Dustin Hansley
Founder, Hans-Made Research
Open to what is next

Partnerships, research collaborations, and large-scale blind tests.

Send us what you are working on. We reply to real enquiries quickly.

Talk to the founder →
Measured blind · third party · reproducible
R² 0.95
E-tongue correlation · r 0.97
RCFTR (U-Manitoba) ran our from-structure bitterness on their Alpha MOS ASTREE, blind, n=20. Above the instrument’s own 0.94 standard curve. No statistical fitting.
70.7%
Six-class taste · vs 33.6% null
On 3,517 public tastants (ChemTastesDB), zero training, against a shuffled-label null of 33.6%.
shuffle → chance
Every channel clears its null
Shuffle the labels and the signal collapses. That is the test that it is physics, not fitting.
01The proof, not our word

A third party’s own instrument, scored cold.

E-tongue taste mapdossier output · sensor-array PCA
Sensor-array PCA taste map from the dossier, the food polyphenols separated by their e-tongue fingerprint into a bitter group and an astringent group.
A real dossier output, not a drawn chart. The engine placed each polyphenol on its e-tongue sensor map from structure alone, blind. Scored against the University of Manitoba lab measurement it correlated at R² 0.95 (r 0.97, n 20), reproduced live on the running platform.
Predicted from structure vs measured Alpha MOS ASTREE bitterness. Eight reference standards plus twelve polyphenol dilutions (naringin, quercetin, curcumin, chlorogenic acid). R² 0.95, Pearson r 0.97, MAE 1.10, dose direction 4/4. Blind, third party: Richardson Centre for Food Technology and Research, University of Manitoba. Source: signed RCFTR report, reproduced live 2026-08-03.

It reads the structure and computes, from first principles, how the molecule lands on each instrument and each taste receptor. No training data. And it names the structural reason for every call.

The lab held the reference values, ran our from-structure bitterness on their own e-tongue, and scored us cold. The correlation sits above the instrument’s own calibration curve, so the read is genuinely predictive, not a fit to the answer key.

See a real dossier →

02Called blind, cold, from SMILES

Six off-notes it had never seen.

Named from structure, then checked against the GC-O bench. The reference labels were withheld until after the call.

Molecule
Predicted character (structure-only)
vs GC-O
2,4,6-trichloroanisole
musty / cork taint
✓ matches the bench
Why

The chlorinated methoxybenzene scaffold is the classic cork-taint signature the nose reads as musty, detectable at parts-per-trillion.

Geosmin
earthy / beetroot
✓ matches the bench
Why

A bicyclic terpenoid alcohol, the same earthy compound the nose picks out from soil and beetroot at trace levels.

4-ethylphenol
barnyard (Brett)
✓ matches the bench
Why

A volatile phenol from the Brettanomyces pathway; the para-ethyl phenol group reads as barnyard and horsey.

Guaiacol
smoky / phenolic
✓ matches the bench
Why

An ortho-methoxy phenol, the core smoke-and-clove marker in lignin pyrolysis and phenolic taint.

1-octen-3-one
metallic / mushroom
✓ matches the bench
Why

An eight-carbon vinyl ketone from lipid oxidation, read as metallic and mushroom even in trace amounts.

Dimethyl trisulfide
sulfurous / cabbage
✓ matches the bench
Why

A small polysulfide; trisulfides read as sulfurous and cooked-cabbage, a marker of staling and over-processing.

A correct blind call is a real prediction, not a lookup. The engine was handed a structure with no name and returned the character; only afterwards was it checked against gas-chromatography-olfactometry.

03How it works

One submission in. The whole bench out.

01

Submit a structure

A SMILES, a sequence, or an InChI. A novel compound works exactly like a known one.

02

The engine reads it

One deterministic pass places it on every instrument at once. No training set, no lookup.

03

You get a dossier

A branded PDF, a results CSV, and a SHA-256 seal you can verify yourself, the same day.

Deterministic. Offline. The same structure always returns the same numbers.

04Method, not accuracy

Four things a SMILES model cannot do.

Trained machine learning is already accurate, so we do not compete on accuracy. We do the four things a model fit to SMILES cannot.

01

Activity cliffs

One atom flips sweet to bitter (sucrose to sucrose octaacetate). We get the cliff right where similarity models fail.

Why

Sucrose is sweet; its fully acetylated cousin, sucrose octaacetate, is intensely bitter. A model that scores by graph similarity sees near-identical molecules and calls them alike. Reading the structure from first principles catches the flip.

02

A named mechanism

Every call states the structural reason it fired, the OECD (Q)SAR standard, auditable, not a black box.

Why

A reviewer can read which group fired and why a call was made, the same white-box standard the OECD asks of a (Q)SAR, instead of trusting an opaque score.

03

Zero training data

Nothing is fit to labels, so it reads novel chemistry no model has seen. A good blind score cannot be leakage.

Why

With nothing fit to the answer key, a correct blind call cannot be memorised training data. Shuffle the labels and the score collapses to chance, the control we publish.

04

Inverse design

Run it backwards, from a target taste to the structures that hit it. Marked as a pilot deliverable.

Why

Because the read is a computation rather than a lookup, it can run in reverse, from a target profile toward candidate structures. This is offered as a pilot deliverable, not claimed as wired today.

The activity cliff, in glass

R- vs S-carvone · one stereobond flipped
O H₃C CH₂ CH₃ mirror
R-(-)-carvone · wedge
spearmint, cooling · TRPM8
S-(+)-carvone · hash
caraway, warm-spicy · TRPV1
Same 2-D graph, one stereobond flipped. Both calls correct, blind, against an identical neutral name. Graph and QSAR taste models are blind to this: enantiomers are graph-identical. Scope: this is class-wired to terpenoid ketones, not a claim of universal chirality.
05It was never just taste

The whole bench, from the sequence.

Nineteen virtual instruments, each rendered in its real instrument’s native format.

AThermal & structural

DSC denaturationTGA / DMACD secondary structureFTIR / Ramanfluorescence

BMass, size & charge

LC-MS intact mass + CCSUV-PDA ε280SEC-MALS / DLS / SAXSzeta / isoelectric pointAUCNMR

CFunction & nutrition

gelation & rheologyemulsificationwater / oil holdingsolubilityDIAASallergen family

DTaste, aroma & shelf-life

primary taste calloff-note culpritranked masking pathwaysaroma driversoxidative stability
Quantitative descriptive analysis radar with seven axes, sweet, sour, salty, bitter, umami, spicy and aroma, plotted from structure.
QDA sensory radar, computed from structure alone. Seven axes: sweet, sour, salty, bitter, umami, spicy, aroma. One of the taste-and-aroma reads returned per input.

Every per-input figure is computed live from your molecule. No canned charts, nothing left blank.

06Benchmarks, with the misses left in

Match, don’t beat.

We match the trained state-of-the-art at zero training. The edge is breadth across the whole bench, not per-axis superiority. And an honest model tells you where it is weak.

Where it excels
2.7°C
Melt temperature, in-family (MAE). Cross-family widens to 6.55°C, n=22, rank ρ 0.83.
70.7%
Six-class taste on 3,517 tastants, against a 33.6% shuffled null.
0.68
Umami call, Matthews correlation. Structure-only and zero-training, on held-out peptides (iUmami UMP442, n=88).
0 wrong
Allergen family across 120 WHO/IUIS food allergens: zero wrong-family calls, it abstains rather than mis-assign. About 91% correct on the plant seed-storage families that dominate them.
ρ 0.93
Gel-strength rank within a uniform processing panel.
Where it declines or misses

Pea vicilin reads about +8°C high on a live blind run, a known pure-vicilin versus convicilin gap. A model tuned to the answer key would not miss like this; a first-principles one does, honestly, and flags it.

Gluten and collagen are refused, not faked. Fibrous and disordered folds fall outside the lane, so the engine declines rather than guess. A calibrated refusal is a feature a surrogate cannot copy.

07Why a screen pays

The value is the cycles you avoid.

Flavor and fragrance is a large market, and taste is the wall in it. Spend the bench only on the candidates worth confirming.

$32-41B
Flavors & fragrances (2024). Taste-modulation is projected to grow to $13.9B by one industry estimate.
$85-175K
and 6 to 12 months, the cost of a single new-product-development attempt.
#1 barrier
Taste is the top wall to plant-based adoption. Pea and soy off-notes must be masked.
Screen 5,
not 50
The leverage is the bench rounds you never have to run. Narrow the set before a single assay.

The unit of value is one purification, sensory panel, or e-tongue session avoided. The losers die in silico.

Computational screening cut time and cost by 25 to 50 percent through R&D (BCG and Wellcome, Nature 2023). Cited as a pharma drug-discovery analogue, not our own figure. For context on scale: meeting the FDA’s voluntary sodium targets alone was modeled at $16.6B over ten years (Milbank Quarterly 2019), a cost of reformulation, not a market size.

08Why you can trust the number

Sealed, named, honest, and yours.

Sealed

SHA-256 integrity seal

Every dossier carries a seal. Change one atom and it changes; re-run and it reproduces.

Named

A mechanism, not a score

Every call states its mechanism in plain flavor-science you can audit, not a black box.

Honest

A band, or a refusal

In-domain it ships a value with a ± band. Genuinely out-of-scope inputs are refused, not guessed.

Yours alone

Your structures stay yours

Proprietary structures run on our side. You receive the results, never the method.

● Re-run = byte-identical The taste blind reproduced three times, bit for bit, under its SHA-256 seal. Reproducibility is the proof it is a computation, not a sample from a model. 9f2a…e7c1 → 9f2a…e7c1
09Built for your lane

One engine. Seven ways in.

beachhead
Precision fermentation

The release panel from the construct.

Predict identity, mass, stability and allergen from the gene, before the bioreactor.

Explore →
Plant protein

Functionality, predicted not measured.

The functional and sensory surface from sequence, before the batch.

Explore →
Masking & sugar reduction

Which masker, and why it works.

Rank the bitter-blocker and name the mechanism, before the panel.

Explore →
Allergen & novel food

The cross-reactivity screen.

The structure-in family and cross-reactivity read EFSA asks for, pre-synthesis.

Explore →
Reformulation

The drop-in, predicted.

The plant replacement and the formulation-change effect, before the trial-and-error.

Explore →
Instrument labs

Predict your instrument suite from a sequence.

The DSC, DLS, CD, LC-MS and zeta reads, computed blind, before the run.

Explore →
The standard

The benchmark the field is missing.

Food-protein functionality has no open benchmark. We compute it from sequence, zero training.

Explore →
Start a blind pilot

Send us the lot that’s fighting you.

Data withheld. We call it cold, from structure. You score us against your own data.

Blind & holdout-scored Same-day Verifiable, line by line