Skip to content
Company

We build computational biology tools in the open

Inflexa is software for computational biology, from preclinical research through translational decisions: an open-source engine you run on your own machine, Inflexa Pro on top of it, and Inflexa Cloud for teams. Every finding carries a signed record of how it was made. We think the tools science runs on should be readable by the scientists who depend on them.

Our story

Why we built this.

We spent years building computational biology software used at biopharma organizations. That work put us close to translational and discovery teams across pharma, biotech, and academia. We saw the same pattern everywhere.

Most assets that move from preclinical to clinical fail. The warning signs are usually there in the data, but nobody has time to find them. Teams handle the routine parts: QC, normalization, basic stats. The bottleneck is the deeper work: comparing methods head to head, checking whether a biomarker holds across cohorts, following a weak signal across three datasets to see if it is real or noise. That work takes weeks of expert time. Most of it never gets done.

The signals that kill an asset in phase 2 are usually visible earlier, if someone has time to look.

So we built Inflexa to do that work. It runs more of the analysis than a person has time for: more methods, more comparisons, more follow-up. It generates hypotheses, then tests them against the data and against the literature. When datasets are large and time is short, humans miss things.

The output isn't a black box, and neither is the software. Every result links back to the data, the code, the method, and the model that produced it, and every citation in it is checked against PubMed and the other registries before it is printed. And the tool that produced it is open under Apache 2.0. You can read it, change it, run it on a machine we never see, and keep using it whatever happens to us.

A scientific result you cannot inspect is a result you are taking on trust. We would rather you didn't have to.
Beliefs

What we believe.

  • Rigor over speed

    Running more methods and cross-validating results takes longer than running one pipeline. When a real decision rests on the answer, the extra time is worth it.

    • Confounder-adjusted DEcompleted,
    • IPTW sensitivitycompleted,
    • GSEA, concordant onlycompleted,

    688 pathways hold after the weighting

    partially confirmed

    Concordant between the severity-adjusted and the weighted rankings, and none changes direction.

    High confidenceT2S1
  • Every finding is traceable

    Every finding links back to the data, the code version, the method, and the model that produced it.

    inflexa · prov

    $ inflexa prov lineage output/pathways.csv

    output/pathways.csv sha256:4c1f…9ab2

    gseapy_prerank.py exit 0

    output/de_joint_adj.csv sha256:8e07…13cd

    limma_trend_de.R exit 0

    output/log2tpm_filtered.csv sha256:a2d9…77e0

    qc_audit.R exit 0

    GSE109142 source

    model anthropic/claude-opus-5 · step T2S1

    signed ed25519:9f3c…a71b

    $ inflexa prov verify

    ok · chain intact · signature valid

  • Scientists stay in control

    The system writes the first draft. You edit and approve. Nothing runs without your sign-off, and because the source is open, nothing runs that you cannot read.

    Approvedinflexa store link limma fgsea --lang r

The steps and the finding are the PROTECT responder session's, as on the home page. The record's hashes and file names are placeholders.

Questions?

We're happy to talk about how Inflexa works or whether it's a fit for your research.