Work / FLkit

FLkit

The know-how to start, gathered in one place

A study across centers that keep their own data needs lawyers, engineers, clinicians and data stewards to work together. What each of them needs to know already exists, but it is spread over framework documentation, legal templates, glossaries and the people who have done it once, and it is hard to find your way through the first time. FLkit gathers it into one open guide.

Question
What does a team need to know before it can start a study across centers?
Constraint
Lawyers, engineers, clinicians and data stewards each need a different part of it, at a different moment.
What I built
One open guide. I led its design, including the template and its elements, coordinated the core group and the working group, and I am first author of the paper on it. I also maintain it.
What changed
A public guide, written with 29 contributors and supported by ELIXIR Belgium.

The decisionEach role gets its own way in, instead of one reading order for everyone, because a clinician and an engineer need a different first page.

Live since 2024, paper 2026. 29 contributors.

On this page
  1. Why teams stop before they start
  2. How the guide is organized
  3. What we built
  4. What it does not show yet
  5. Links and paper

What it produced

39 pages, 11 ways in

Four life cycle stages and seven worked project stories.

Visit FLkit. Pirmani, Vermeulen et al., arXiv 2606.23500, 2026.

Why teams stop before they start

A team that has never run a study like this cannot tell which piece it needs first: the clinician does not know which legal template matters, and the engineer does not know what an ethics board will ask.

If a tool needs an engineer to get started, most small teams never start, and the groups that need this most rarely have a systems team. I judge my own work by whether a small group can run it without me in the room.

How the guide is organized

We ordered the material the way the decisions actually happen.

ELIXIR. The European network for life science data.

The format comes from the ELIXIR research data management kit, which many research teams already know.

The same pieces, before and after.

What we built

  • Four life cycle stages: governance, infrastructure, data wrangling and analysis, across eight sections.
  • A way in for each role, so nobody has to read all of it.
  • Seven worked project stories, from multiple sclerosis, inflammatory bowel disease, genomics and brain computer interfaces.
  • Open. Anyone can propose a change or a new page.

Pick your role, and see where FLkit sends you first.

  1. Governancedefine use case, consent, ethical framing
  2. Infrastructurenot the main focus for this role
  3. Wranglingclinical data curation, harmonisation
  4. Analysisinterpretation, evaluation, impact analysis

Open this role in FLkit

In FLkit's own words, from its role pages. Five roles are shown here; FLkit has eleven.

The detailStatus and sources

What it does not show yet

The numbers above describe what is in FLkit. They do not show how many teams use it, or whether it helps them start faster. That has not been measured yet.

Links and paper

Back to the first story

When a registry was not allowed to send its data COVID-19 and multiple sclerosis

Let's talk.

If your question depends on real world health data, I probably want to hear about it.

Email me

TurkishMerhabaAzeri, my mother tongue, as written in Urmia: xoş gəldin, welcomeخوش گلدینAzerbaijani, as written in AzerbaijanSalamPersian, the language of school: dorudدرودEnglishHelloDutch, as said in FlandersDagyou
A greeting in each language I speak, where it lives.