Method & data
How we work with your data.
What we can analyse, how a project is framed, and how your data are handled. Specific terms are agreed in writing for each project.
Data we work with
- Bulk RNA-seq: raw counts, normalised expression (TPM, CPM) or FASTQ on request
- Clinical or experimental metadata: treatment, response criterion, follow-up, covariates
- Public cohorts: GEO, TCGA and other open repositories, used under their terms
- Single-cell or spatial data: discussed case by case
Formats
- CSV, TSV or Excel tables; R or Python objects on request
- One row per sample, with a data dictionary for metadata
- Pseudonymised identifiers only — no names, dates of birth or direct identifiers
Analysis principles
- A written analysis plan before results are produced
- Quality control and documented preprocessing
- Robustness checks: resampling, covariates, alternative definitions
- Exploratory and confirmatory findings clearly separated
- Reproducible code and versioned results kept for the project
Confidentiality
- A confidentiality agreement can be signed before any exchange of data
- Data used only for the agreed project, never reused without written consent
- Storage, transfer and deletion or return of data defined in the project agreement
- Results belong to the terms agreed in the contract
What we do not do
- We do not receive identifiable patient data through this website or by plain email
- We do not provide diagnoses, treatment or dosing recommendations
- We do not present research findings as validated clinical tests
First message: please describe your data in general terms only. We will agree on a secure transfer method before any data are shared.
Discuss a pilot like this
Tell us about your candidate, data and question. We will come back with a scoping proposal.