Case studies

Work, and the figures it produced

Four engagement formats, with the type of data and analytics each one delivers. Client names are withheld; the figures illustrate the deliverable rather than a published client result.

01 — Industrial biotech

Screening twelve strains on a four-run budget

Candidates evaluated
12
Runs required
4
Predicted yield spread
2.3×

The problem

Twelve candidate production strains and funding for four shake-flask runs. The team needed a defensible way to choose which four, and what to conclude if none performed.

The work

Yield and growth parameters were estimated from published data, then a sensitivity run identified which variables the ranking actually depended on. A staged decision tree was built so early readouts would eliminate candidates instead of merely ranking them.

Delivered

A four-run screening plan with predicted spread between candidates, the two variables to hold constant, and a pre-agreed stopping rule.

02 — Process economics

Cutting substrate per gram without touching titre

Substrate per gram
−18% (model)
Titre held at
Baseline
Profiles compared
3

The problem

Carbon source dominated variable cost, but the team could not change product concentration without losing downstream yield.

The work

A mass balance over the fed-batch was rebuilt from the run logs, and three feed profiles were modelled against the same titre constraint. Sensitivity to feed timing was tested rather than assumed.

Delivered

A revised feed profile, the balance sheet behind it, and the operating window in which the saving holds.

03 — Molecular design

Why an expression construct produced nothing

Changes flagged
3
Frame-shift risk
Confirmed
Cassette variants
2

The problem

A cassette that worked in one host gave no detectable product in another. Two rounds of troubleshooting had replaced reagents but not read the sequence.

The work

The full construct was re-annotated: reading frame, ribosome binding site spacing, restriction sites, and codon usage against the new host. Each candidate cause was ranked by how cheaply it could be tested.

Delivered

An annotated sequence map with three flagged changes, ranked by likelihood and test cost, and a rebuilt cassette with corrected spacing.

04 — Analytical documentation

An assay two operators could reproduce

Protocol steps
8
Acceptance criteria
3
Variables isolated
5

The problem

A colorimetric assay drifted between operators and days. Results were being discarded rather than diagnosed, and nobody could say which step was responsible.

The work

The method was written as an SOP with every timing, temperature and dilution made explicit, acceptance criteria defined before the next run, and a control placed at each step that could vary.

Delivered

An eight-step SOP, three acceptance criteria, and an operator checklist to be signed at each batch.

Your project's first deliverable can be scoped from a two-paragraph brief.

Describe your project →