← All enterprise use cases

11 / Monoclonal antibodies

Digital twins: testing gas-delivery choices before a process handoff

A receiving-site workflow for examining oxygen supply, carbon-dioxide removal and controller behavior before selecting a bioreactor study.

Proposed enterprise workflow · Source-linked analysis · Expert review

Conceptual illustration of bioreactor evidence and a scientific review workflow

THE ENTERPRISE OBJECTIVE

A defined question.
A shared evidence basis.

Help process development, MSAT and engineering choose a gas-delivery scenario for receiving-site confirmation, with the model assumptions and unresolved risks visible.

Start with the receiving decision

A hypothetical antibody program is moving into a vessel with a different sparger and gas-control configuration. The original dissolved-oxygen setpoint is available, but the receiving team needs to understand whether its equipment can support the culture over the full run. Copying a setpoint does not reproduce the physical environment.

The decision owner

Process development and MSAT own the scientific interpretation, with engineering, analytical and quality colleagues reviewing evidence relevant to their responsibilities.

Why the decision matters

The business dependency is the receiving-site engineering slot. A study that merely repeats the reference setpoints may consume that slot without resolving whether a gas-delivery constraint is equipment-driven or culture-driven. The team needs a reasoned comparison of candidate strategies, the observations that could overturn the preferred option and a clear account of what the receiving vessel can physically execute.

Which candidate gas strategy deserves a physical engineering run?

Connect gas-flow histories, agitation, oxygen enrichment, pressure, dissolved oxygen, cell density and available off-gas measurements to the same batch timeline. Retain calibration records, controller settings and periods of sensor uncertainty. Distinguish measured transfer characteristics from values estimated using correlations.

The evidence boundary

Keep observations, derived quantities and model outputs distinguishable. Missing evidence remains an explicit gap in the review record, with an owner and a proposed way to resolve it.

HOW BIOMEDANA FITS

From source context
to a reviewable decision.

  1. 01

    Reconstruct the reference

    Bio Hub organizes the source run, instrument context and equipment record so reviewers can trace every selected input.

  2. 02

    Define the model boundary

    Scope a project-specific Twin assessment for gas transfer and culture demand. Document unmeasured quantities, fitted parameters and unsupported outputs.

  3. 03

    Compare feasible scenarios

    Evaluate candidate gas and agitation strategies against the receiving equipment envelope. Treat recommendations as study proposals, with controller implementation assessed separately.

  4. 04

    Design the receiving study

    MSAT and engineering agree the measurements, stop conditions and review responsibilities before testing the selected candidate.

Challenge the explanation before using the estimate

Begin with oxygen demand and supply, then examine carbon-dioxide removal and equipment limits separately. A culture can meet a dissolved-oxygen target while another gas-related constraint remains unresolved. Test whether the model distinguishes transfer limitations from changes in cellular demand; multiple parameter combinations may explain the same observed oxygen trajectory.

Evaluation for this decision

Hold out complete runs and examine the periods of highest demand, controller transitions and recovery after disturbances. Compare prediction intervals with observations at those periods rather than reporting only an average fit. A useful scenario includes the possibility that the model cannot rank two candidates confidently.

Make the model assumptions inspectable

A useful starting representation separates the dissolved-oxygen inventory, transfer into the liquid and cellular consumption. Working volume, gas composition, pressure and transfer behavior shape the supply side; viable biomass and cell-specific demand shape the consumption side. A fitted consumption term should not quietly compensate for an incorrect transfer assumption. Where the measurements cannot distinguish them, retain alternative parameter sets and show how that ambiguity changes the scenario ranking.

Make the next action explicit

Review finding / Next action
Review findingNext action
A feasible candidate has an adequate comparison basisSelect the receiving study with agreed gas measurements, equipment constraints and stop conditions.
Supply and demand cannot be distinguishedObtain independent transfer or respiratory evidence before assigning the excursion to a mechanism.
The candidate requires unsupported actuationRevise the candidate or scope the controller and equipment work before testing it.

How to evaluate value

Track time to a reviewed study proposal, unresolved equipment assumptions and agreement with the receiving run. Assess relevant quality measurements independently of an oxygen prediction.

THE WORKING ARTIFACT

What your team reviews.

A gas-delivery study pack with source runs, equipment assumptions, candidate trajectories, uncertainty, controller constraints and an owned confirmation protocol.

Scope and interpretation

This proposed workflow supports engineering-study selection. Model-assisted assessment does not establish a deployed closed-loop controller, product comparability or verified receiving-site performance.

This scenario describes a proposed enterprise workflow. Deployment-specific integration, review and model behavior require verification. Business and process impact are evaluated against an agreed baseline and measured outcomes.

Explore the modality ↗

Explore BioMedAna Twin ↗

The evidence needed to make this comparison useful

Build the evidence set around the receiving question. The following review map identifies what each source contributes and the uncertainty it is intended to resolve.

Evidence layer / What to retain / Question it helps answer
Evidence layerWhat to retainQuestion it helps answer
Culture demandViable biomass, available respiratory measurements, growth phase and feed context.Is the excursion associated with a different biological demand profile?
Transfer and equipmentWorking volume, sparger, agitation, pressure and the basis of transfer estimates.Does the receiving configuration support the candidate throughout the run?
Control executionCascade order, actuator limits, deadbands and recorded mode changes.Does the model assume an action the actual controller cannot deliver?
Independent responsesCarbon-dioxide evidence, viability and relevant product-quality measurements.What would reveal an unacceptable trade-off despite an improved oxygen trace?

Two plausible explanations for the same oxygen excursion

Consider an illustrative receiving run in which dissolved oxygen dips as viable biomass rises. The sending team sees a familiar oxygen-demand peak and favors earlier enrichment. The receiving engineer sees a controller reaching its configured limit and suspects a different actuator envelope. Both explanations are plausible from the oxygen trace alone. The investigation preserves those alternatives instead of naming the first association as the root cause.

The first comparison reconstructs what the controller requested, what the gas train delivered and what the culture was doing at the same time. If delivered flow differs from the recorded request, the review turns toward instrumentation or execution. If delivery is consistent but estimated demand differs substantially from the reference cohort, the biological assumptions deserve a closer look. If neither can be established, the missing measurement becomes part of the study objective.

Now compare two proposed strategies. One relies more on enrichment; the other changes the balance of total flow and agitation within the equipment envelope. The team examines oxygen supply, carbon-dioxide observations, controller saturation and independent culture responses. A candidate with an attractive predicted oxygen trace can remain unsuitable when its carbon-dioxide behavior is uncharacterized or its required actuation is unavailable. The decision record should state that reason explicitly.

A supported conclusion might therefore be to test a candidate while adding the measurement that separates supply from demand. It might also be to perform equipment characterization before a culture run. These are conditional paths in a proposed investigation, not reported findings or an operating recipe.

A staged path from the first review to repeatable use

Four delivery stages

  1. 01

    Retrospective reconstruction

    Start with an existing transfer question and reproduce the source timeline. Review which gas and culture signals are observed, estimated or absent before building scenarios.

  2. 02

    Offline scenario review

    Compare a small set of feasible candidates using a frozen model version. Have the receiving engineer challenge every equipment assumption and the scientist challenge the demand representation.

  3. 03

    Physical confirmation

    Carry the selected question into an authorized engineering study. Predeclare the measurements that would support, weaken or invalidate the scenario interpretation.

  4. 04

    Controlled reuse

    Add the receiving evidence to the project record after review. Reassess the supported vessel and operating region before reusing the model for another handoff.

What travels with the decision

The receiving packet should identify the reference runs, model version, equipment configuration and uncertainty that matter to the selected scenario. It should state who evaluates oxygen behavior, who reviews independent quality evidence and who can stop or revise the study. Preserve rejected candidates and their reasons so a later team does not reopen an already resolved equipment question without new evidence.

Questions to settle before starting

Why is matching the dissolved-oxygen setpoint insufficient?

A setpoint describes a control target. It does not specify gas-transfer capacity, controller response, cellular demand or carbon-dioxide removal. The receiving study needs evidence about those relationships under its own equipment conditions.

Can the model choose a final gas recipe?

The proposed use is to prioritize a study and explain its assumptions. Final operating instructions need the relevant physical evidence, process review and execution controls; a scenario ranking alone does not supply those.

What if off-gas or carbon-dioxide data are unavailable?

Show the limitation in the candidate comparison. Use sensitivity analysis to assess its consequence and decide whether targeted measurement or equipment characterization is needed before advancing.

Scientific background and further reading

These references provide background for the scientific methods and review considerations discussed here. The scenario, delivery plan and decision examples are proposed BioMedAna workflows; the publications do not establish BioMedAna deployment or performance.

START WITH YOUR PROCESS QUESTION

Turn the evidence
into a reviewable next step.

Bring your program context, available evidence and the decision your team needs to make.