From microbiome analysis to predictive insight

The challenge

Eagle Genomics was developing a platform combining data processing, statistical analysis and network science. Its ambition was to help research organisations understand biological relationships, predict effects and support defensible product claims.

I joined to turn an inherited hypothesis-testing prototype into a brand-compliant application. My remit expanded into product direction, scientific workflow design and leadership of product designers and data-science prototypers.

The challenge was to turn advanced scientific capabilities into a coherent product that scientists could use with confidence.

What I uncovered

The prototype covered one happy path. The application needed to support different hypothesis types, experimental conditions and analytical choices within a much wider scientific journey.

Working with microbiologists, bioinformaticians and data scientists revealed the judgement behind those choices. Scientists needed to understand whether their data was suitable, which methods were appropriate and what conclusions the results could support. Removing code from the workflow would not, by itself, make the analysis trustworthy.

The longer-term hypergraph ambition also lacked a clear connection between the technology and the purposes it would serve. Discovery revealed three distinct needs: developing products towards a known outcome, finding new uses for known microbes and exploring relationships without a predetermined result. Each required a different starting point and journey.

My approach

I wanted to connect the platform around the journey from a scientific question to a defensible insight, embedding specialist knowledge where scientists needed to make decisions.

That meant making methods, assumptions and confidence visible throughout the analysis. It also meant defining how findings could become reusable organisational knowledge: retaining their evidence and provenance so later research could build on or challenge them.

I used concepts and working prototypes to make that direction tangible, while bringing customer value, scientific requirements and technical feasibility into decisions about what to pursue next.

What I changed

Embedded scientific expertise in guided analysis

I translated specialist knowledge into a guided Microbiome Statistical Analysis experience, helping microbiologists move from their data and experimental questions towards appropriate analyses and interpretable results.

Working with scientific specialists, I designed the journeys, validation, defaults and visible assumptions needed to support those decisions. The designs connected analytical choices to visual results and recommended hypothesis tests, while showing where expert scrutiny was still required.

This commitment, coupled with her attention to detail, and deep understanding of IT architecture, consistently ensured exceptional product designs that fully met scientific needs.

Turned the hypergraph ambition into a tangible product direction

I defined how scientists could explore connected evidence for different research and product-development goals, then created the initial concepts.

The direction showed how findings from separate studies could become evidence-bearing relationships, retaining their provenance and statistical confidence as knowledge evolved.

Designers under my leadership developed the high-fidelity designs, which the prototyping team turned into a functioning front end for exploring networks containing thousands of microbes. This gave stakeholders a concrete experience to evaluate and exposed the scientific value, evidence requirements and performance that still needed validation.

Aligned the roadmap with customer value and delivery feasibility

I made the choices between the longer-term vision and a deliverable next step explicit. Customer journeys and prototypes helped stakeholders assess the value of connected knowledge alongside dependencies in data preparation, performance and support.

The team prioritised a focused improvement to the existing Exploration experience, rather than committing to the complete hypergraph proposition before those dependencies were resolved.

The ability to influence all stakeholders, including c-suite with her design vision and execution is well evidenced.

Built design and prototyping capability

I moved design upstream into discovery and product definition, bringing scientific specialists and customers into the work before requirements were fixed.

I established shared design foundations and linked, version-controlled Figma flows, giving designers, product and engineers a common reference for the intended experience. I also brought data-science prototyping into the design team, connecting concept development with the scientific and technical expertise needed to explore it.

Outcomes

The work delivered a functioning network-exploration front end, giving stakeholders a concrete basis for evaluating the experience and identifying dependencies before committing further.

The guided statistical-analysis designs received positive customer feedback. Design became involved earlier in product decisions, with scientific expertise, customer research and prototyping contributing before requirements were fixed.

The delivered statistical-analysis MVP did not reliably realise the intended experience. Eagle entered administration before dependable adoption, scientific outcomes or commercial impact could be established.