Plant Protein Replacement at Scale
Abstract
The plant-based food company partnered with Apoha to innovate two products, one live product and one in development. On the live product they looked at replacing a key ingredient to improve product quality and cost for consumers, Apoha identified two effective protein substitutes from 15 candidates, which one was then selected for use. The second project involved using a high performance but also high cost ingredient, the plant-based food company wanted to find an innovative way to ensure they can provide consumers with best possible product, Apoha found that a formulation with 5–15% inclusion of this high performance plant protein matched full-inclusion performance. Apoha's work ran alongside the plant-based food company's existing R&D without disrupting it, accelerating innovation that would otherwise have taken months into two weeks, with initial results within 2 days after receiving samples.
In Collaboration with:
UK Plant-based Food Company
How To Read This
Investigation 1 - Protein Substitution: For live product improvement, screening 15 candidate proteins to identify a replacement for a core ingredient of a product that was already on supermarket shelves.
Investigation 2 - Formulation Efficiency: Identifying the minimum effective inclusion of a high-cost functional ingredient in a product in development. Keep cost as low as possible for consumers whilst providing the best possible quality.
About Liquid State Intelligence™
A new class of data: behavourial data
Liquid State Intelligence™ places a substance under controlled stress and reads how it responds, capturing a unique behavioural fingerprint that reveals how something truly acts, not just what it's made of.
Liquid State Intelligence™ generates a new class of behavioural data that reveals how materials truly compare, giving teams a deeper basis for confident decisions faster that conventional characterisation cannot provide.
The complexity of molecular behaviour within formulations and how processing techniques changes this makes product development slow, costly, risky and hard to predict, conventional techniques can't capture enough of this behaviour at commercial speed. Liquid State Intelligence™ captures complex behavioural data across ingredients, formulations and processing conditions that conventional techniques cannot, enabling faster decisions, greater control over development, and significant reduction in R&D timelines and cost.
In complex beverage formulations, single-component characterisation provides valuable insight into individual ingredients. However, emergent interfacial and viscoelastic behaviours arising from multi-molecular interactions require a whole-system view. Liquid State Intelligence™ builds on established approaches, going further by measuring the system as a whole and resolving behaviours that current methods leave uncharacterised.
Customer Background
The plant-based food company creates hyper-realistic meat alternatives for meat-eaters.
The plant-based food company excels at innovation and providing an elevated consumer experience. The plant-based food company faces formulation decisions both in active development and in live products already on shelf. The plant-based food company is committed to delivering the best possible results for its consumers, and looks for the smartest, most efficient ways to bring innovation to market, without slowing down the rest of its development pipeline.
Plant protein supplier specifications typically cover purity and pH, useful as a starting point, but they say little about functional behaviour. Viscosity, solubility, and gelling can be measured independently, but how a protein performs within a complex formulation matrix, and under real manufacturing conditions, is difficult to predict. Plant proteins respond differently to different environments, and their behaviour in a product is nonlinear, meaning small changes in formulation or process can produce disproportionate effects on the final result.
The Challenge
Protein Substitution
The plant-based food company wanted to replace a key ingredient in a product already on supermarket shelves as efficiently as possible, ideally in weeks rather than months. As with any live product, the plant-based food company needed complete confidence the change would hold up at full production scale, with no compromise to the quality and experience consumers already knew.
This could have been a long-term project that would have taken resources away from another product, but the plant-based food company wanted to do this as effectively as possible to deliver more than one improvement across their range. The plant-based food company needed to make this change with zero disruption to a product already on shelf, protecting both batch quality and the experience consumers already knew. Conventional ingredient data gave a useful starting point but not enough confidence to act on alone. With 15 candidate proteins to evaluate and no time for extended physical trials, the plant-based food company needed a faster, more confident way to identify the closest functional match.
The Second Challenge
Formulation Efficiency
The plant-based food company wanted to find the minimum amount of a high-cost ingredient needed to deliver the same product performance, without slowing down the rest of the innovation in the project.
The ingredient was high-performing but high-cost, and reducing its use without compromising quality was a priority for keeping the product competitively priced . The relationship between inclusion level, cost, and performance was nonlinear and hard to predict, requiring complex formulation design rather than a simple like-for-like reduction. The plant-based food company wanted a precise, evidence-based answer: the lowest inclusion level that delivered full performance while also protecting consumer cost.
Plant protein supplier specifications, covering parameters such as purity, pH, fat and fibre content, provide a useful compositional baseline but reveal little about functional behaviour. Conventional physicochemical measurements such as solubility, viscosity, and gelling can supplement this, but they do not capture how a protein behaves within a complex formulation matrix or under real manufacturing conditions. Ingredient interactions in plant-based formulations are nonlinear and process-dependent, meaning performance cannot be reliably predicted from single-property measurements alone.
How Characterisation Was Applied
The plant-based food company had 15 candidate proteins to choose from and wanted a fast, effective way to know which would actually work in their formulation at scale. Apoha used a two-stage approach: Stage 1 benchmarked all 15 candidates against the gold-standard protein to identify the closest behavioural matches; Stage 2 tested the top 4 candidates within a partial formulation matrix containing the key functional ingredients, capturing how each protein behaves in the presence of the components that most influence product performance, giving the plant-based food company an evidence-based shortlist in weeks.
Wanting to avoid a long-term trial, the plant-based food company had conducted conventional physicochemical testing on the candidates, but wanted the highest possible confidence in the chosen protein replacement, to fully protect product quality at launch.
Each candidate protein was measured at controlled concentration, generating a high-dimensional behavioural fingerprint per sample. Interfacial behaviour provided a first-pass behavioural comparison across all 15 candidates. PCA was then applied to the fingerprint data, enabling visual clustering and quantitative similarity scoring relative to the gold-standard reference. Surface pressure was captured as a separate readout. The top 4 candidates in behavioural similarity, which also matched with internal characterisation, were re-measured within the actual base formulation matrix, capturing how each protein behaves in the presence of the other formulation components, where interaction effects determine real functional performance.
Results, Outcomes and Key Data
Some proteins looked similar on the surface pressure measurement, but deeper analysis revealed which ones truly matched the target behaviour. Apoha identified two effective alternatives. The plant-based food company then considered the cost and availability of each, and the candidate that best matched the target behaviour also proved the most viable commercially. This change was used in the product on the market.
Figure 1 shows surface pressure measurements across all 15 candidates, with a dotted line indicating the gold-standard reference value. Several candidates sit close to the reference level, providing a useful first-pass filter, but surface pressure alone was not sufficient to shortlist with confidence. Surface pressure, a well-understood physical quantity, was used to corroborate Apoha's waveform-derived PCA findings, providing independent verification of the behavioural similarity results. Figure 2 shows PCA of the full fingerprint data provided the additional resolution needed, identifying two candidates whose combined behavioural profile most closely matched the gold standard, which would have been difficult to distinguish by surface pressure alone.
Plant proteins are known to be highly variable in both composition and behaviour, even within the same protein type. Figure 2 shows this variability clearly, with the four candidates displaying a range of profiles in PCA space. Despite this, a clear overlap in behaviour between S1 and the gold standard S2 is visible, with S3 showing partial overlap as a secondary candidate.
Combining the PCA results with the surface pressure measurements, two candidates emerged as the closest matches to the gold standard. These results were also consistent with the physicochemical data the plant-based food company had measured independently. Both were taken forward for physical trials. One candidate was selected, successfully trialled at scale, and incorporated into the recipe, with the product now on the market.
Figure 1 shows surface pressure measurements across all 15 candidates, with a dotted line indicating the gold-standard reference value. Several candidates sit close to the reference level, providing a useful first-pass filter, but surface pressure alone was not sufficient to shortlist with confidence. Surface pressure, a well-understood physical quantity, was used to corroborate Apoha's waveform-derived PCA findings, providing independent verification of the behavioural similarity results. Figure 2 shows PCA of the full fingerprint data provided the additional resolution needed, identifying two candidates whose combined behavioural profile most closely matched the gold standard, which would have been difficult to distinguish by surface pressure alone.
Plant proteins are known to be highly variable in both composition and behaviour, even within the same protein type. Figure 2 shows this variability clearly, with the four candidates displaying a range of profiles in PCA space. Despite this, a clear overlap in behaviour between S1 and the gold standard S2 is visible, with S3 showing partial overlap as a secondary candidate.
Combining the PCA results with the surface pressure measurements, two candidates emerged as the closest matches to the gold standard. These results were also consistent with the physicochemical data the plant-based food company had measured independently. Both were taken forward for physical trials. One candidate was selected, successfully trialled at scale, and incorporated into the recipe, with the product now on the market.
.png)
.png)
How Characterisation Was Applied
The plant-based food company needed to find the right amount of a high-cost ingredient to include in a new product, too much would increase consumer costs, too little could compromise quality. Apoha tested the formulation across a range of inclusion levels to find the point where a small amount delivered the same performance as much more.
Apoha ran a structured study mapping how formulation performance changed across inclusion levels of the functional ingredient within the base matrix. This identified the minimum threshold needed to maintain product quality, enabling the plant-based food company to make a confident, evidence-based decision in weeks rather than months.
Apoha's Liquid State Intelligence was used to characterise formulation behaviour across a range of inclusion levels of the high-functionality ingredient within the base matrix. Rather than measuring the ingredient in isolation, the analysis focused on interaction effects between the functional ingredient and the formulation, capturing the emergent, nonlinear behaviour that governs real product performance. PCA of the high-dimensional fingerprint data was used to identify the inclusion regimes where behaviour shifted most significantly, mapping the inclusion-response relationship and identifying where lower inclusion levels produced behaviour overlapping with the high-inclusion reference. This approach was able to compare complex, highly similar materials with high sensitivity.
Results, Outcomes & Key Data
This identified a clear sweet spot where a low inclusion level achieved performance overlapping with much higher-cost formulations. The plant-based food company could therefore use the expensive ingredient strategically rather than maximally, preserving product specifications while significantly improving cost efficiency. Rather than months of trial-and-error or an educated guess that could have introduced significant doubt into the project, the decision was made with confidence and without the time needed for a physical trial.
Formulations with higher proportions of the high-functionality protein cluster in the upper region of PCA space, while the 100:0 formulation sits distinctly separate. Notably, the 25:75 ratio shows the greatest spread from the 100:0 reference, suggesting that at this inclusion level the high-functionality protein may behave differently in the presence of the lower-functionality protein, though the mechanism is not fully understood and this observation requires further investigation. Formulations at 5:95, 10:90, and 15:85 show the closest behavioural overlap with the 100:0 reference, indicating that as little as 5–15% high-functionality protein achieves comparable performance to the pure high-functionality formulation.
.png)
Next Steps
This work gives the plant-based food company a faster, evidence-based way to make formulation decisions on future products, whether replacing ingredients or optimising cost, without slowing down the rest of their innovation pipeline. Apoha continues to expand its work across plant-based proteins with a range of partners.