
How does a protein find one gene among twenty thousand?
Zinc fingers scan DNA by sliding and hopping. We time each step with single-molecule imaging and model the search.
Our zinc fingers read eighteen letters of DNA. That is enough to find a single address in a genome of three billion.
Each question is a project, a grant and at least one argument at lab meeting. We publish what we find, including when the answer is no.

Zinc fingers scan DNA by sliding and hopping. We time each step with single-molecule imaging and model the search.

Our activators recruit the cell’s own machinery. No breaks, no scars, and the edit fades when the protein does.

The same switch works in one cell and stalls in its neighbour. We map chromatin state across whole organoids to find out why.
Drag or scroll sideways. Each slide is a project page with the methods, the people and the data behind it.
All research
A light-sheet pipeline that images whole organoids every twenty minutes without bleaching them.

Designs that the cell clears on a schedule, so an experiment can have a defined start and end.

A designed protein that recognises one eighteen-letter stretch of DNA, and the measurements that show how tightly it holds.

An open, low-cost screening pipeline that lets one person characterise a thousand protein designs in a week.

Fourteen structures showing how designed proteins meet packaged DNA, and why the angle of approach matters.

Tracking single molecules as they slide and hop along DNA, and a two-number model that predicts how long the search takes.
Hover the plot to read values. Every panel is redrawn from the numbers in our papers, not a screenshot.
Eight recent papers. Every one links to the journal and, where we have it, the raw data.
All publicationsFerreira A., Weber J., Maraval I.
Okafor R., Lindqvist S., Maraval I.
Weber J., Nair P., Maraval I.
Okafor R., Brandt T., Maraval I.
A group of physicists, chemists and biologists who all had to learn each other’s vocabulary.
Meet the lab
Structural biology and protein design. Trained as a physicist.

Cryo-EM of designed proteins on packaged DNA.

Computational design and molecular dynamics.

Single-molecule tracking in living cells.
The short version of a long CV. The full PDF has every grant, student and committee.
Group leader since 2019, tenured in 2024.
Structural biology · 11 peopleStarted the lab with one microscope and two students.
ERC Starting GrantFirst zinc-finger activator solved on DNA by cryo-EM.
EMBO FellowshipHow transcription factors find their sites, measured one molecule at a time.
Thesis prizeMoved from optics to biology by way of a microscope I built.
With distinctionThree talks coming up and a list of the last few. Slides on request, always.
All talksShort posts about method, failure and how we run the lab.
All posts
Nine hundred designs did nothing. The pattern in which ones did nothing turned out to be the most useful result of the year.

Three questions we ask before a project starts, and the one we ask six months in.

Why every new person gets instrument time in their first week, before they know what they are doing.
You have a system, we have the structures. Most of our papers start with an email like yours.
PhD and postdoc places open twice a year. Read how we work, then tell us what you want to learn.
Seminars, keynotes and school visits. I say yes to one outreach event a month.