☰

BRAIN Biotech AG

Connecting multi-site fermentation onto one data path

Industry
Industrial biotechnology (specialty enzymes and microbial fermentation)
Headquarters
Zwingenberg, Hesse, Germany
Public information as of
February 2026

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

Strategic priorities

BRAIN Biotech has set out a 'Vision 100' to reach EUR 100 million in revenue for its Biocatalysts segment within five years, with a 15 percent adjusted EBITDA target. The FY 2024/25 figures put the group at EUR 49.6 million of revenue and an adjusted EBITDA just below zero, so the gap to the target is the operating problem the rest of the group is now organised around.

The route through the gap runs through a multi-site industrial footprint: R&D at Zwingenberg in Germany, the Biocatalysts manufacturing hub at Cardiff in the UK with fermenters up to 63,000 litres, and a new greenfield production site at Nieuwkuijk in the Netherlands whose construction is under way. The Capital Markets Day framing makes the implication explicit: 'With Industry 4.0, all machinery, control systems, and IT systems will be connected horizontally and vertically throughout the organisation.'

The supporting spine is uneven. The CFO has publicly cited 'misalignment between operations and financial reports' as something that 'obscures visibility, stalls innovation and delays decisions,' and the group runs an advanced voluntary sustainability reporting programme aligned with the VSME standard across 16 UN Sustainable Development Goals. The 2024 royalty monetisation with Royalty Pharma brought in up to EUR 128.88 million and the Akribion Therapeutics licensing deal up to EUR 92.3 million in milestones, putting the cash in place to fund the new facility and the upgrades.

Lab discovery already has a proprietary engine: the MetXtra database and AI-supported prediction models are used to discover and optimise enzymes. The handoff between that engine and the production fermenters, though, is still largely manual: scientists transfer process parameters by hand into production systems, which is the work environment that has to change for the next scale-up to be repeatable.

Challenges we see

  • Operations Integration

    Aligning operational and financial reporting across the group

    The CFO has publicly identified 'misalignment between operations and financial reports' as a barrier to visibility and decision-making. The group operates R&D at Zwingenberg, manufacturing at Cardiff and a new production site in Nieuwkuijk, with the analyst note from the Capital Markets Day noting that the EUR 128.88 million royalty monetisation with Royalty Pharma and the EUR 92.3 million Akribion Therapeutics licensing deal are now funding CAPEX across sites.

    Where the same batch is described once in the engineering record and once in the financial system, the two views have to be reconciled by hand. Letting the operational system write the same events into the financial record at source collapses the reconciliation into a single source of truth.

  • Digital Integration

    Bridging Zwingenberg discovery and Cardiff production on a shared data path

    The MetXtra database and AI-supported prediction models drive enzyme discovery at Zwingenberg, while the production fermenters at Cardiff carry the work into industrial scale. The 6M 2024/25 management statement notes that the Biocatalysts segment runs against a 'sales and cost base mismatch' that the new IT/OT backbone is meant to close.

    Where the lab system and the fermenter control system do not share an ontology, every scale-up transfers parameters by hand. Mapping both onto a common assay, organism and process-parameter model lets the discovery output flow into the production record automatically.

  • Digital Manufacturing

    Bringing existing and new fermenters onto one communication backbone

    The Cardiff manufacturing site runs through Biocatalysts Ltd with industrial-scale fermenters up to 63,000 litres, and the group is constructing a new greenfield production site in Nieuwkuijk, Netherlands. The 2024 Capital Markets Day statement names a standardised OPC UA (Open Platform Communications Unified Architecture) backbone as the path to horizontal and vertical connectivity.

    Where fermenters and skids come from different vendors and the new site is being commissioned from scratch, agreeing the protocol and information model up-front means the data path is a property of the design rather than an integration project after handover.

  • Operations Manufacturing

    Carrying scale-up learning from benchtop to 63,000-litre fermenters

    Scaling enzymes from 0.5 litre desktop to 63,000 litre industrial fermenters is described as 'long and expensive,' and the Biocatalysts site notes that 'developing enzyme production processes is often reliant on a single partner.' Antifoam management and metabolite dynamics at scale both shape yield at the new volumetrics.

    Where the next litre is run on operating envelope rather than on the prior litre's data, the development cycle re-derives what an instrumented fermenter would have shown. Treating the benchtop and the production fermenter as one data path with the same parameter model makes the handoff a query instead of a relaunch.

  • Compliance Regulatory

    Producing sustainability and GxP evidence from one set of records

    The group operates under food safety scrutiny for its Biocatalysts products, applies the VSME standard for voluntary sustainability reporting across 16 UN Sustainable Development Goals, and has BioIncubator programmes being prepared for FDA/EMA GxP transitions, including the Akribion Therapeutics CRISPR-Cas platform.

    Where energy, batch, sample and operator records each live in their own system, assembling a sustainability disclosure or a GxP evidence packet means a manual cross-reference. Keeping per-batch records with their lineage intact from the start makes the same record serve sustainability, regulatory and GxP questions.

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. Reading operations and financials from one group-wide data path

    The CFO has publicly described 'misalignment between operations and financial reports' as something that 'obscures visibility, stalls innovation and delays decisions,' and the group operates across Zwingenberg, Cardiff, Nieuwkuijk, WeissBioTech, AnalytiCon Discovery and Biosun Biochemicals.

    An ontology-based industrial data platform with OPC UA standardisation at the equipment boundary lets operational events feed the same data model that the financial system reads, so the operational and financial views stop being a manual reconciliation.

    • Capital Markets Day Statement, BRAIN Biotech, December 2024
    • BRAIN Biotech AG IR Management Statement 6M 2024/25

  2. Instrumenting the Cardiff fermenters alongside the Nieuwkuijk greenfield

    The Cardiff site runs industrial-scale fermenters up to 63,000 litres with manual data extraction, and the new Nieuwkuijk facility is under construction without a defined digital nervous system. Foam management at the new volumetrics is a known scale-up risk.

    Retrofitting the Cardiff fermenters with online sensors and computer-vision foam control, and specifying the same data contract in the Nieuwkuijk procurement, gives one data path for the running site and the new one.

    • Biocatalysts Ltd, fermentation services overview
    • BRAIN Biotech CMD, December 2024

  3. Designing the Nieuwkuijk data path before the equipment is bought

    The new production facility in Nieuwkuijk is under construction. Equipment selection and network design decisions taken now determine for the next decade whether plant data is reachable from enterprise systems.

    Setting the OPC UA information models, network segmentation and equipment data requirements before procurement means interoperability arrives with the machines, instead of becoming an integration project after handover.

    • BRAIN Biotech CMD, December 2024, Nieuwkuijk greenfield
    • BRAIN Biotech AG IR Management Statement 6M 2024/25

  4. Linking MetXtra predictions to confirmable bench experiments

    The MetXtra database and AI-supported prediction models drive enzyme discovery at Zwingenberg, and the AI-to-production handoff is described in the research as a place where scientists hesitate to act on model output without further confirmation.

    A scientist-facing dashboard that exposes prediction confidence, the underlying MetXtra evidence and the controlled experiment used to test it lets researchers move from AI output to confirmed result in a single, traceable workflow.

    • BRAIN Biotech AG, R&D Spotlight, MetXtra
    • BRAIN Biotech CMD, December 2024

  5. Turning scale-up and sustainability data into a recurring report

    Tracking 16 UN Sustainable Development Goals across multiple jurisdictions, and standing up GxP evidence for BioIncubator assets, both depend on stitching energy, batch and operator records into a single document. Each cycle is largely a manual compile.

    Narrow AI agents can draft scale-up review notes and sustainability disclosures from the per-batch record, check completeness against the template, and surface every controlled document a standards change touches, with a named reviewer approving each output.

    • BRAIN Biotech, sustainability statement, VSME standard
    • BRAIN Biotech AG, FY 2024/25 results

What we'd propose

  • Enterprise AI

    Industrial data platform for group-wide visibility

    An ontology-based industrial data platform that unifies data streams from all BRAIN sites against OPC UA at the equipment boundary, so the operational and financial records read from the same events and the group's IT/OT layer is set once and reused across sites.

    • Group-wide ontology

      One agreed set of terms

      Define equipment, batch, organism, sample, assay and process parameter as explicit entities with agreed relationships, so operational and financial systems can read from the same model rather than from two parallel ones.

    • OPC UA backbone at the equipment boundary

      Machine data into a standard

      Specify OPC UA at the equipment boundary so controllers, fermenters and skids from different vendors publish into a documented, vendor-neutral layer rather than each one staying inside its own controller.

    • Self-service dashboards for operations and finance

      The same data, two views

      Expose the model through Grafana dashboards and a retrieval layer so the engineering team reads the operational view and the executive team reads the financial view from the same events, with no manual cross-reference in between.

    • The operational and financial views stop being a manual reconciliation and become one read of the same events.
    • The same backbone works for the running Cardiff site and the new Nieuwkuijk site, so the second site is not a separate integration project.
    • CFO and CEO work from the same data source, with the lineage from process value to financial figure auditable.

    We've done this before

    See how we do Enterprise AI
  • Digital CDMO

    Fermenter retrofit for Cardiff and the data path for Nieuwkuijk

    A retrofit programme on the Cardiff fermenters (online sensors, computer-vision foam control, anomaly detection) and a design package for the Nieuwkuijk procurement that specifies the same data contract, so the running site and the new site sit on the same data path.

    • Online sensor retrofit on existing fermenters

      Live data on legacy equipment

      Install non-invasive sensors on existing 10 kL and 63,000 L fermenters to capture temperature, pressure, pH and dissolved oxygen in real time, with the OPC UA interface already in place for the platform above.

    • Computer vision foam control

      Catching overflow before it happens

      Deploy camera systems with detection algorithms on the bioreactors to flag foam buildup before it triggers a process failure, and feed the events into the same batch record as the other process parameters.

    • Nieuwkuijk equipment data contract

      Interoperability built in at procurement

      Write the OPC UA information models and the data publishing requirements into the Nieuwkuijk procurement documents, so interoperability is a purchase condition rather than an integration project after handover.

    • The Cardiff fermenters and the Nieuwkuijk greenfield publish into the same data model, so the running site and the new site are one system.
    • Foam events and other deviations surface against the batch that is running, not as a footnote in the post-batch report.
    • Equipment data is an asset of the operating company, not a feature of whichever vendor supplied the controller.

    We've done this before

    See how we do Digital CDMO
  • Digital Lab

    Lab system integration around MetXtra and the bench

    An integration of the MetXtra discovery platform with the bench instruments and the bench data capture, so a prediction from MetXtra is paired with the experiment designed to confirm it, and the confirmation writes back into the same record the discovery came from.

    • MetXtra to bench instrument data flow

      Predict and test against one record

      Wire the MetXtra database to the bench instruments so a proposed enzyme variant and the experiment designed to characterise it share a single record, with the prediction confidence and the measured result visible together.

    • Confidence and lineage in the scientist UI

      A scientist reads the reasoning, not a number

      Surface prediction confidence, the data sources used to derive it, and the controlled experiment that confirmed or refuted it, so researchers can review the AI output on the same screen as the wet-lab evidence.

    • Validation workflow tracking

      Confirmation steps tracked end-to-end

      Track the chain of validation experiments, version numbers and outcomes against the parent prediction, so the audit trail from a published enzyme to its supporting evidence is queryable instead of reconstructed.

    • Researchers work from a single record that joins prediction, experiment and outcome, instead of separate databases.
    • Confidence and lineage are visible to the scientist, not buried in a model file.
    • The bridge from MetXtra to the production fermenters is documented in the same record, so the scale-up reads the discovery history upstream.

    We've done this before

    See how we do Digital Lab
  • Agents

    AI agents for scale-up review and sustainability reporting

    Narrow, reviewable agents that take the recurring document work: drafting scale-up review notes from the per-batch record, assembling sustainability disclosures against the VSME template, and finding every controlled document a standards change touches. A named reviewer approves every output.

    • Drafting from source records

      First drafts from the data

      Generate the first draft of a scale-up review note or a sustainability disclosure directly from the per-batch record, so the author edits and judges rather than assembles the document from the operational systems.

    • Template and completeness checking

      Gaps found before review

      Check a submitted document against the VSME template, the GxP evidence list or the group's own checklist, and return 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 UN SDG reporting scope, a GxP rule or a VSME standard 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.

    • Review queues move faster because documents arrive complete and traceable to the per-batch record.
    • 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

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.

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 BRAIN Biotech AG's own published ambition implies — not a perfect score.

Source: A4BEE analysis of public sources
IT/OT connectivity 25 → 85
The Capital Markets Day statement names horizontal and vertical connectivity as the goal, and the Cardiff site is described as running with outdated hardware design. The Nieuwkuijk greenfield is the chance to set the new baseline.
Data integration 30 → 90
The CFO has publicly identified 'misalignment between operations and financial reports' as a barrier to visibility, and the discovery-to-production handoff is described as a manual transfer. The shape of the work is acknowledged; the systems are not yet joined.
Process automation 35 → 80
Scale-up from 0.5 L to 63,000 L currently depends on operator experience and on a single partner for the development work. The data path is being defined; the automated control architecture is being scoped.
Real-time analytics 25 → 85
Shop-floor KPIs are disconnected from executive dashboards, and AI predictions are not yet trusted in production. The internal data side is ahead; the production-floorside is the work to do.
Regulatory and sustainability 45 → 90
Food safety and the VSME-standard sustainability reporting are both maintained, and BioIncubator programmes are being prepared for GxP transitions. The compliance scope is clear; the underlying record layer is manual.
Change readiness 40 → 75
The research identifies a 'digital hesitancy' around AI tools among scientists and a 'War for Talent' that makes efficiency gains a strategic priority. The leadership framing is supportive; the cultural adoption is the next step.

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 BRAIN Biotech AG, 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].