A bespoke cloud LIMS built to absorb pandemic testing volume
A custom cloud LIMS validates each genetic sample scan in 250ms, cut the scan error rate to 0.5%, and saved 70 seconds of manual reporting per sample — proven across 6M+ samples as Oncologica scaled from hundreds to tens of thousands of PCR tests a month.
The challenge
- COVID-19 pushed Oncologica's PCR testing from a few hundred to tens of thousands of samples a month, and the lab's infrastructure could not keep up.
- Manual reporting was slow, labour-intensive, and error-prone — unsustainable at the new volume.
- A vast amount of sensitive health data had to be handled securely and kept compliant with strict standards.
What we did
- We built a cloud LIMS that validates genetic sample scans in real time, replacing the manual reporting that had become the bottleneck.
- The system runs as microservices on Google App Engine, so infrastructure security sits with the provider and the team can focus on the application.
- A headless, REST-API architecture separates the core engine from the front end, letting mobile apps and third-party systems reach core functions securely.
- The interface was built in Vue.js, and the platform meets HIPAA, ISO/IEC 27001, and HITRUST CSF with AES-256 encryption and SSL-secured transfers.
Results
- 6M+samples validated in production
- 0.5%scan error rate
- 250msto validate a sample
- 70ssaved per sample in manual reporting
“The platform absorbed a volume we could never have handled on paper, and it did it without letting error rates climb.”
Head of Laboratory, Oncologica
The context
Oncologica is one of the top five diagnostic laboratories in the UK, running PCR tests and COVID sequencing for the public and for NHS hospitals. A team of around 650 people covers lab science, operations, and support.
When COVID-19 hit, testing demand did not grow gradually. It jumped from a few hundred samples a month to tens of thousands. The lab's manual reporting was slow, labour-intensive, and error-prone, and it was never designed for that volume. On top of the throughput problem, a large amount of sensitive health data now had to be handled securely and kept compliant.
What we built
The core of the work was a cloud Laboratory Information Management System that validates each genetic sample scan in real time, replacing the manual reporting that had become the bottleneck.
- Microservices on Google App Engine, so infrastructure security sits with the provider and the team can focus on the application.
- A headless, REST-API architecture that separates the core engine from the front end, letting mobile apps and third-party systems reach core functions securely.
- A Vue.js interface for lab operators to run validation and reporting.
- Compliance with HIPAA, ISO/IEC 27001, and HITRUST CSF, with AES-256 encryption and SSL-secured transfers.
How it works in production
A scan comes in, the system validates the sample information against the record, and the result is reported without a person retyping it. What used to be manual checking and reporting now happens in a fraction of a second, and the same path handles a handful of samples or a full batch without changing.
The outcome
The system has been proven on more than six million samples. The scan error rate settled at 0.5%, validation runs in 250ms, and roughly 70 seconds of manual reporting is saved per sample, which adds up to about 1.8 hours per testing batch. Interactive KPI calculation feeds a continuous improvement plan, so the lab keeps tuning the process rather than firefighting it.
§ 03
Have a workflow like this?
Book a 30-minute scoping call: we map one workflow, the modules it needs, and the hours a system would return.