☰

Aphea.Bio NV

Pilot-plant data that survives an EFSA submission

Industry
Agricultural Biotechnology (Microbial Biopesticides and Biostimulants)
Headquarters
Ghent, Belgium
Public information as of
February 2026

A4BEE prepared this analysis from publicly available sources. It reflects our own reading of Aphea.Bio NV's published strategy and is not endorsed by, or produced in cooperation with, Aphea.Bio NV. Company website

Strategic priorities

Aphea.Bio emerged from VIB (Vlaams Instituut voor Biotechnologie) in 2017 and has grown from 10 employees at founding to over 65 by 2025, anchored at the Ghent Bioaccelerator where it screens more than 100,000 microbial strains on its APEXbio™ platform. The capital trajectory crossed €70 million with the July 2023 Series C, and a follow-on VLAIO Powerhouse grant in September 2025 funds the move from discovery lead into commercial product.

The immediate operational shift is the new Ghent pilot plant, where fermentation moves from 1 L laboratory bioreactors into 500–5,000 L vessels. Two flagship biofungicides, VALORIA™ and VIRTUOSA™, and the INITIA maize biostimulant sit ahead of EPA (US Environmental Protection Agency) and EFSA (European Food Safety Authority) registration, with US and Brazilian market launches sequenced into the 2026–2027 window. Field-trial data from more than 225 sites in Poland, the US and Brazil feeds both the regulatory dossier and the agronomic case.

The data and automation work has to be in place before those dossiers land. APEXbio™ generates terabytes of microbiome-mapping and greenhouse phenotyping data, but it sits in functional silos disconnected from the pilot plant, and field-trial records are still partly captured on paper and in Excel. Pilot-scale fermentation control recipes, instrument data capture, and a single ontology for assay, batch and lot are the building blocks that determine whether an EFSA reviewer reads clean evidence or a reconstruction.

The strategic frame is sustainability: Aphea.Bio holds B-Corp certification, is targeting Ecovadis Silver, and has committed to a 10 percent reduction in Scope 1 and 2 emissions by 2027. The Bill & Melinda Gates Foundation joined the Series C round, signalling an accessibility and affordability mandate for low-income geographies. Each of those commitments turns into evidence the company has to produce, and the producing is increasingly a software problem rather than a documentation problem.

Challenges we see

  • Operations Manufacturing

    Moving fermentation recipes from 1 L to 500–5,000 L without rework

    Aphea.Bio is moving promising microbial candidates from 1 L laboratory bioreactors into 500–5,000 L pilot vessels, where fluid dynamics, aeration and nutrient distribution behave differently. The pilot plant was inaugurated in late 2023 and the first commercial launches are sequenced into 2026–2027.

    Where control recipes and automation logic are built under launch pressure, the practical expectation shifts from reworking fermentation late in the cycle to specifying the modular control architecture before each strain reaches the pilot floor.

  • Digital Integration

    Joining APEXbio™ discovery data to pilot plant operations data

    The APEXbio™ platform generates terabytes of data from microbiome mapping and automated greenhouses, but the data sits within functional units without a unified connection to pilot plant operations. Field trials span more than 225 sites in Poland, the US and Brazil.

    Where discovery data and operations data are described in different shapes, the question of whether the two estates load against one model determines whether downstream analytics ask one question or reconcile per report.

  • Operations Operations

    Replacing manual laboratory steps with image-based classification

    Microbial cell counting and the identification of colony infections remain manual microscopy tasks, even though the platform already runs AI-based disease scoring on plant phenotypes. The benchtop work draws on the time of trained laboratory staff.

    Where cell counting and colony classification are visual tasks performed under a microscope, computer-vision models can take the monotonous step out while the trained scientists spend their time on the interpretation that follows.

  • Compliance Regulatory

    Producing regulatory-grade batch evidence across paper and Excel workflows

    EPA and EFSA submissions for VALORIA™ and VIRTUOSA™ require thousands of pages of toxicity, ecotoxicology and efficacy data, alongside B-Corp and Ecovadis reporting. Field-trial records still move partly through paper and Excel, and the regulator expects clean, contextualised evidence from source to scientist.

    Where regulatory submissions have to be assembled from records that were partly captured by hand, the practical shift is from producing evidence at the end of the cycle to capturing it directly from the instrument or field device as the work happens.

  • Digital Integration

    Connecting diverse laboratory and pilot equipment to one data path

    The Bioaccelerator and the new pilot plant use equipment from multiple vendors speaking different protocols (RS-232, Profibus, vendor-specific), and the diversity is set to grow as commercial-scale capacity is added.

    Where equipment is described by vendor protocol rather than by an open standard, the question of whether the resulting data is reachable from enterprise systems is settled by the integration architecture chosen at procurement, not by retrofit later.

Opportunities, by urgency and business impact

Each bubble is one opportunity, numbered to match the list below. Further right means it bites sooner; higher means a bigger effect on the business. A bigger bubble means a bigger implementation effort.

Source: A4BEE analysis of public sources

  1. Standardising pilot-scale fermentation automation around MTP

    Moving promising microbial candidates from laboratory to pilot vessels requires building control recipes and automation logic under launch pressure, and different strains ask for different module configurations.

    An MTP (Module Type Package) architecture, the NAMUR VDI/VDE 2658 standard for modular process equipment, lets each fermentation skid describe its control interface once, so reconfiguration for a new strain is module substitution rather than code rewriting.

    • Aphea.Bio deep research, §1 (Ghent pilot plant inauguration late 2023)
    • VLAIO Powerhouse grant announcement, September 2025

  2. Putting discovery and pilot data on one ontology-based lakehouse

    APEXbio™ microbiome mapping, automated greenhouse phenotyping and pilot plant operations each describe their data in their own shape, and the discovery-to-pilot boundary is where the gap is widest.

    An ontology-based data platform maps physical fermentation parameters, microbial strain, assay, lot and field trial onto one set of entities, so the discovery data and the operations data load against the same model instead of being reconciled downstream.

    • Aphea.Bio Series C announcement, July 2023
    • Aphea.Bio deep research, §1 (APEXbio™ platform)

  3. Capturing field trial records as data for the regulatory dossier

    Field trials across more than 225 sites in Poland, the US and Brazil generate the agronomic evidence behind the EPA and EFSA filings, and part of that evidence is still captured on paper or in Excel.

    A paperless laboratory execution layer captures results directly from field instruments and benchtop devices, with audit trails and ALCOA+ (Attributable, Legible, Contemporaneous, Original, Accurate, plus Complete, Consistent, Enduring, Available) compliance built in, so the regulator reads data that was captured once rather than reconstructed for the submission.

    • Aphea.Bio deep research, §1 (225+ field sites)
    • VALORIA™ and VIRTUOSA™ EPA registration programme, 2025

  4. Automating cell counting and colony classification under the microscope

    Manual counting of microbial cells and identification of colony infections under microscopes draws on the time of trained laboratory staff, and the work is monotonous and prone to human drift.

    An image-classification model trained on the company's own microscopy data counts cells and flags infections in seconds, returning the scientist's time to interpretation and freeing bench throughput.

    • Aphea.Bio deep research, §1 (APEXbio™ high-throughput screening)

  5. Connecting diverse laboratory and pilot equipment through OPC UA

    Equipment from multiple vendors using RS-232, Profibus and vendor-specific protocols sits at the Bioaccelerator and the new pilot plant, and the protocols don't natively speak to one another.

    A vendor-agnostic integration layer with OPC UA (Open Platform Communications Unified Architecture) gateways, MTP-aware modules and secure IT/OT segmentation brings 30 or more devices onto a single data path without replacing the equipment.

    • Aphea.Bio deep research, §1 (Bioaccelerator expansion)
    • NAMUR VDI/VDE 2658 MTP standard

What we'd propose

  • Digital CDMO

    MTP-based modular automation for the Ghent pilot plant

    An MTP (Module Type Package) reference architecture for the pilot plant: each fermentation skid exposes its control interface through NAMUR VDI/VDE 2658, so adding a new strain or swapping a module is reconfiguration rather than code rewriting.

    • MTP library for fermentation skids

      Reusable control modules

      Build an MTP-compliant library of control modules for bioreactor, feed, harvest and CIP (Clean-In-Place) skids, with object-oriented interfaces so a new strain recipe selects the modules rather than writing PLC (Programmable Logic Controller) code from scratch.

    • Process orchestration layer

      Coordinating the modules

      Add a Process Orchestration Layer (POL) that schedules and coordinates the MTP-enabled Process Equipment Assemblies (PEAs) so a multi-step fermentation run is choreographed at the plant level rather than driven from each skid individually.

    • OPC UA backbone

      Vendor-neutral data path

      Deploy OPC UA as the communication backbone so fermentation data and control commands travel over one documented, vendor-neutral protocol that the enterprise side can read without bespoke integration per vendor.

    • Adding a new microbial strain to the pilot plant becomes module reconfiguration rather than a control-system rewrite.
    • Recipe authoring moves from PLC programming to selecting and parameterising modules, shortening the path from lab lead to pilot batch.
    • The same MTP library and POL pattern can be reused at commercial scale, where the cost of bespoke automation rises steeply.

    We've done this before

    See how we do Digital CDMO
  • Enterprise AI

    Ontology-based data platform across discovery, trials and pilot

    An ontology-driven data platform that defines the entities Aphea.Bio shares across discovery, field trials and the pilot plant — microbial strain, assay, lot, field site, batch — and loads APEXbio™ phenotyping, field-trial and pilot operations data against that single model.

    • Shared microbial and process ontology

      One agreed set of terms

      Define microbial strain, fermentation parameter, assay, specimen, lot, field site and batch as explicit entities with documented relationships, so a question written once returns comparable answers across discovery and operations.

    • Streaming and batch pipelines

      Both kinds of data on one platform

      Implement streaming pipelines for live pilot plant and benchtop instrument data alongside batch pipelines for greenhouse phenotyping and field-trial records, with schema validation at the boundary so malformed data fails loudly instead of landing silently.

    • Analytics and retrieval layer

      Questions answered without IT tickets

      Expose the model through dashboards and a retrieval layer so discovery, regulatory and pilot teams can query the combined data set directly, including the historical runs that anchor each new recipe.

    • Integration work is done once against a shared model instead of once per point-to-point interface.
    • Discovery leads and pilot runs are queryable in the same place, which is what the strain-to-batch traceability story depends on.
    • New data sources attach to the ontology rather than triggering another migration.

    We've done this before

    See how we do Enterprise AI
  • Digital Lab

    Paperless laboratory and field-trial execution layer

    A laboratory execution system (LES) for benchtop work and field-trial records that captures data directly from instruments and field devices, with audit trails and ALCOA+ compliance built in so the regulatory submission reads data that was captured once.

    • Instrument and field-device capture

      Data captured at source

      Connect balances, plate readers, environmental sensors and field data loggers so results are captured with instrument identity, method version, location and timestamp attached, instead of being transcribed from a screen or a notebook.

    • GxP-aligned execution layer

      ALCOA+ at the point of capture

      Build ALCOA+ (Attributable, Legible, Contemporaneous, Original, Accurate, plus Complete, Consistent, Enduring, Available) data integrity, electronic signature, versioning and audit-trail handling into the execution layer so a regulator-facing record stands on its own.

    • Secure IT/OT data flow

      Validated path to enterprise

      Design a high-availability architecture with secure gateways between the laboratory and field network and the enterprise systems, so validated, tamper-evident data reaches the regulatory data platform without manual transfer.

    • Field and benchtop data are captured once and reused for the dossier, instead of being reconstructed for each submission.
    • Audit risk on EPA and EFSA filings falls because the regulator-facing record is the same record the scientist worked from.
    • B-Corp and Ecovadis reporting pulls from the same evidence base rather than a parallel sustainability spreadsheet.

    We've done this before

    See how we do Digital Lab
  • Agents

    AI agents for EPA, EFSA and sustainability document work

    Narrow, reviewable agents that draft regulatory and sustainability documents from source records, check submissions against the relevant template before human review, and find every controlled document a standards change touches. A named reviewer approves every output.

    • Drafting from source records

      First drafts from system data

      Generate the first draft of EFSA and EPA submission sections, B-Corp and Ecovadis reports, and internal periodic reviews directly from the laboratory and field records, so the author edits and judges rather than assembles.

    • Template and completeness checking

      Gaps found before review

      Check a draft against the relevant EPA, EFSA, B-Corp or Ecovadis template and the company's own checklist, returning missing or inconsistent sections before the document enters the human review queue.

    • Change impact search across the document set

      Which documents a change touches

      When a regulatory standard, EFSA guidance or sustainability framework changes, retrieve every controlled document that references it and rank them by how directly they are affected, so the update scope is known on day one.

    • Regulatory submission review queues move faster because documents arrive complete.
    • The scope of a standards change is established by search rather than by recollection.
    • Every output is traceable to the source records it came from and signed off by a named reviewer.

    We've done this before

    See how we do Agents
  • Digital Lab

    Vendor-agnostic connectivity for laboratory and pilot equipment

    An integration framework that connects 30 or more diverse laboratory and pilot devices from multiple vendors through OPC UA gateways, custom protocol drivers and MTP-aware modules, so the equipment data reaches the data platform without bespoke integration per device.

    • Custom protocol drivers

      Reading legacy equipment

      Build custom drivers for the RS-232, Profibus and vendor-specific protocols already deployed at the Bioaccelerator, so legacy instruments become reachable as OPC UA endpoints without replacing the hardware.

    • Network segmentation to IEC 62443

      Securing the OT layer

      Define VLAN (Virtual Local Area Network) segmentation and remote-access rules to IEC 62443, the international standard for industrial automation and control system cybersecurity, so the operational technology (OT) network is isolated while validated data flows to the IT (information technology) estate.

    • MTP-aware module wrapper

      Plug-and-play for new skids

      Wrap each new pilot plant skid as an MTP-compliant module so it joins the data platform through the same control interface, regardless of the underlying vendor.

    • Data from existing benchtop and pilot equipment reaches the platform without vendor-by-vendor integration.
    • Security segmentation is set before the next pilot skid arrives, not retrofitted after.
    • Adding a new instrument becomes a configuration step rather than an integration project.

    We've done this before

    See how we do Digital Lab

Where QB Systems fits

Alongside our services we build QB Systems, hardware and software for bioprocess control. QB Systems is a product brand of A4BEE Sp. z o.o.

Applications
Bioreactors

Software-defined control for a bioreactor — a new QB vessel, an upgrade to one you have, or a retrofit of the existing PLC.

Buffer & media preparation

Automated preparation of growth media and process buffers, so a recipe runs the same way every time without fixed infrastructure.

Automated sampling

Automated sampling from 4–18 sources, aseptic-capable and up to 72 hours unattended. Works with any vendor's bioreactor.

Scale
Pilot (50–300 L)

Stainless steel, where QB supplies the control software and integration and a certified partner builds the installation.

Digital maturity: today and target

Scored out of 100 across six dimensions. The target is what Aphea.Bio NV's own published ambition implies — not a perfect score.

Source: A4BEE analysis of public sources
Data Integration 32 → 78
APEXbio™ generates terabytes of phenotyping and microbiome data, but it sits within functional units; the discovery-to-pilot boundary is where the integration gap shows most clearly, and the ontology work is ahead of the company rather than behind it.
Process Automation 38 → 82
AI-based plant disease scoring is in production, but benchtop steps such as cell counting remain manual and pilot plant control recipes are still being assembled under launch pressure; the MTP reference architecture will determine whether each new strain reaches the pilot floor without a control rewrite.
IT/OT Convergence 25 → 75
Equipment at the Bioaccelerator and the new pilot plant speaks RS-232, Profibus and vendor-specific protocols; without an OPC UA backbone the data from these instruments cannot reach the enterprise estate, which is why the integration work has to be specified before procurement.
Regulatory Compliance 42 → 88
EPA and EFSA filings for VALORIA™ and VIRTUOSA™ and B-Corp and Ecovadis reporting all run on the same underlying data; the dossier has to read evidence that was captured once, which is why paper and Excel records are the gap to close first.
Predictive Analytics 22 → 72
Predictive process models are not yet deployed for the pilot plant; the data platform work is the foundation that makes process simulation and strain-comparison analysis possible, and the discovery side is already producing the training data.
Sustainability Evidence 35 → 80
The 10 percent Scope 1 and 2 reduction by 2027 and the Ecovadis Silver target both rely on per-site and ideally per-batch energy and material measurement; without that measurement the sustainability evidence is reported annually rather than managed during the year.

Check this yourself

Our Service Portal has free self-assessments and market comparisons. These are the ones that line up with what we've read above — no sales call required.

Think we've read this right?

Talk to us

Related reading

This is an independent analysis prepared by A4BEE from publicly available information as of February 2026. It reflects A4BEE's own interpretation and opinion, is not affiliated with, endorsed by, or verified with Aphea.Bio NV, and may be incomplete or inaccurate. All company names and trademarks are the property of their respective owners. To request a correction or removal, contact [email protected].