Ovcahub¶
Ovcahub is the system of record for high-grade serous ovarian cancer (HGSC) research: patients, samples, and lab-generated result metadata — with provenance, role-based access to patient identifiers, a global search/discovery layer, a draft-to-verified review workflow for genomic results, and full read/write access from R and Python.
Raw sequencing/imaging files (FASTQ, BAM, WSI) live in EGA and are out of scope here; this system tracks the samples they came from and the results derived from them.
What you can do¶
| If you are… | You can… |
|---|---|
| a lab tech | register samples, upload result CSVs |
| a clinician | everything above plus see patient identifiers (PII) |
| a bioinformatician | browse samples/results, read and push data from the SDK (PII hidden) |
| a postdoc | review pushed results and mark them verified or rejected |
| a superuser | everything — PII, verification, user/audit administration |
The core ideas¶
- Patients carry a study ID. Clinical detail (stage, treatment lines, outcomes) arrives in a
later slice — those columns show
—for now. - Samples have a running
idand a human-readablename(patientid_phase_site), a type (fresh frozen, FFPE, organoid, ascites, plasma), and may be derivatives of other samples. - Results land two ways: as generic ad-hoc observations (a value plus the full source row kept as provenance), or as typed results (somatic/germline variants, copy number, HRD) with real columns and a draft → verified review workflow. Every result is linked to the run (upload) it came from.
- Discovery answers "what data exists?" across the whole corpus with one search box and facets — not just per sample or patient.
- Availability catalog answers "what data exists for this sample / patient?" at a glance.
Continue with Getting started.