Customer story
FYR Diagnostics builds a production-ready multiomic platform on Fuzzball
FYR Diagnostics partnered with CIQ to give its AI-enabled EV-Omics platform a scalable, reproducible analytics foundation, with onboarding cut from days to minutes and up to 100x throughput potential across 8 to 10 pipeline types.
100x
Throughput potential
Across the multiomic pipeline suite
Minutes
To onboard a new hire
Down from multiple days per pipeline
8–10
Pipeline types
Each with distinct requirements, all running consistently
0
Local installs
Validated pipelines run as soon as credentials are issued
The science, the bottleneck, and what changed
How a precision medicine company turned bespoke bioinformatics pipelines into production services its pharmaceutical partners can rely on.
- Who
- FYR Diagnostics, a precision medicine company based in Missoula, Montana
- What they do
- Capture RNA and protein signals from extracellular vesicles in a blood sample through the AI-enabled EV-Omics (EVO) platform, producing multiomic profiles for biomarker discovery, patient stratification, and pharmacodynamic insight in neurology and oncology
- The challenge
- Inconsistent execution environments across languages and machines, multi-day onboarding per pipeline, and AWS-heavy workflows that pulled scientists into infrastructure work
- The solution
- Fuzzball as a container-first production layer: standardized, tracked workflows with centralized logs, data, and execution history
- The results
- Up to 100x throughput potential, onboarding cut from days to minutes, and consistent execution across 8 to 10 pipeline types
“We were a rapidly growing biotech running bioinformatics pipelines on whatever compute resources were available. The scale of data generated is on par with a technology software company, but our core product is a platform that helps pharmaceutical partners accelerate drug discovery and development. To do that effectively and reproducibly, a strong software foundation is essential.”
The challenge: scaling innovation into industry-ready execution
FYR generates data at the scale of a technology company, but its product is a platform that helps pharmaceutical partners accelerate drug discovery. Before Fuzzball, that platform ran on whatever compute was available. Independently installed packages across several languages drifted out of sync between systems, so onboarding a new hire onto a single pipeline could take days and still end in version conflicts. Manual configuration made growth risky, and AWS-heavy workflows demanded engineering expertise that pulled scientists away from the research itself.
FYR needed an analytics foundation that could support its role as a trusted partner without turning its scientific team into an infrastructure team.
The solution: Fuzzball as the production layer
Rather than rebuild internal infrastructure from scratch, FYR implemented Fuzzball as a production-grade layer above its pipelines. CIQ’s container-first approach removed the environment inconsistencies and infrastructure bottlenecks that threatened research timelines, and let the team concentrate on what differentiates its platform: interpreting complex biology and delivering partner-ready insights.
Designed for scientists.
Fuzzball’s interface lets the team run workflows without AWS expertise or long parameter commands.
Container-first by default.
Every execution uses identical package versions. A lab technician or new hire can run the same pipeline the moment they receive credentials, with nothing to install.
Exploration and production, kept apart.
Exploratory work happens outside the platform. Production services run through standardized, tracked workflows with centralized logs, data, and execution history.
“Fuzzball allows us to turn bespoke workflows into stable, reproducible automation. It supports internal programs, academic collaborators, and pharmaceutical partners with consistent, scalable services we can stand behind.”
The impact: from bottleneck to confidence
By removing infrastructure friction, CIQ helped FYR scale its insights responsibly, without compromising scientific integrity.
Faster onboarding
Onboarding once took multiple days per pipeline and often ended in version conflicts. Validated pipelines now run immediately, in minutes instead of days.
Reproducible results
Across 8 to 10 pipeline types with distinct requirements, every execution runs the correct, current configuration, so results match across users.
Focus on interpretation
With consistent pipelines in place, the team’s attention moved from computational logistics to interpreting multiomic data and advancing biological insight.
Ready for scale
Each new omics layer adds dimensionality and data volume. Standardized workflows and simple onboarding let FYR grow sample throughput and headcount with confidence.
Lessons for biotechnology teams
FYR’s experience points to a few things any organization generating technology-scale scientific data should plan for.
Infrastructure is a strategic asset.
Computational infrastructure gets less attention than lab platforms in early-stage funding, yet it becomes the bottleneck as data volume grows. Organizations generating technology-scale data need technology-level infrastructure.
Containers simplify standardization.
Teams without containerization expertise still benefit from a container-based platform that hides the complexity while delivering consistency and reproducibility.
User experience matters in scientific computing.
Graphical interfaces and intuitive workflows lower the barrier for scientists, so they spend their time on research instead of command-line operations.
Turnaround time compounds.
Samples move through many wet lab stages over weeks, and physical processes can only be accelerated so far. Computational analysis is the stage a team fully controls, and it is where Fuzzball gives time back: validated pipelines run on demand, with no setup delays and no reruns caused by environment drift.
Exploration and production need clear boundaries.
Individual exploration and production-quality workflows can coexist when the platform draws an explicit line between them, with tracking and centralization on the production side.
Looking ahead: scaling with confidence
FYR is expanding its neurology and oncology programs, including data showing blood-based detection of adult malignant gliomas with 100% specificity and 89% sensitivity, and Parkinson’s disease biomarker programs to support patient stratification and pharmacodynamic readouts. As partnerships grow and data complexity increases, Fuzzball provides the computational backbone to keep the rigor and reproducibility that regulated environments require while delivering the speed pharmaceutical partners expect.
Ready to scale your computational workflows?
If your organization faces inconsistent environments, lengthy onboarding, or infrastructure bottlenecks that limit growth, these resources show how Fuzzball can help.
The full FYR Diagnostics case study
The complete story as a PDF, from onboarding bottlenecks to production-scale reproducibility.
See how Fuzzball works
Architecture, capabilities, and deployment options for container-first workflow orchestration.
Fuzzball for life sciences
How bioinformatics and drug discovery teams run reproducible pipelines at scale on Fuzzball.
About FYR Diagnostics
FYR Diagnostics is a precision medicine company using extracellular vesicles to reveal RNA and protein biomarkers from blood samples. Through its AI-enabled EV-Omics platform and proprietary SPARCs technology, FYR unravels disease mechanisms to inform diagnostic and treatment options across neurology, oncology, and other therapeutic areas. Based in Missoula, Montana, FYR partners with pharmaceutical companies to improve biomarker discovery, clinical trial design, and patient care.
Get the case study
See the full story of how FYR Diagnostics turned bespoke bioinformatics pipelines into reproducible production services on Fuzzball.