MatriXense Tour

Welcome to

MatriXense

Software that lets you see, and measure, what cells really do in 3D.

  1. 01Why cell biology in 3D is stuck
  2. 02What MatriXense Studio does about it
  3. 03How to bring it into your lab
→to begin
MatriXense

MatriXense Studio

Cells live in 3D.
Now measure them there.

The matrix, the forces and the interactions of cells in 3D. Label-free, automatic, on your own computer.

Meet a fibroblast

Fiona.
Her job: repair.

One of billions of cells inside you, pulling, crawling and talking to each other.

Cut a finger, and she crawls into the wound, grabs the fibres and pulls it closed.

Today we follow her. And measure her world.

Fiona · fibroblast

Why 3D

Cells live in 3D. Most assays don't.

On plastic · 2D
Flat and stuck.Spread on one hard surface, a million times stiffer than tissue.
In a matrix · 3D
Surrounded and pulling.Held by soft fibres on every side, reshaping the matrix as it moves.

The problem

3D gels are easy to make.
Hard to measure.

One well
0images
0 GBof data

12 h, every 10 min · 50 z-slices · 2 channels

Weeks of clicking cells by handEvery cell tracked in 3D, automatically.
Dyes that change how cells behaveLabel-free. Just brightfield.
Pore size from one 2D sliceThe real 3D network, with error bars.

The problem

9 in 10 drug candidates
fail in clinical trials.

Regulators are moving to human models.FDA Modernization Act 2.0 (2022) removed the animal-testing requirement. In 2025 the FDA set out a plan to phase it out.
3D human-cell assays are ready.Measuring them in depth, at scale, has been the bottleneck. Until now.

Sun D. et al., Why 90% of clinical drug development fails and how to improve it. Acta Pharm Sin B 12, 3049 (2022).

Sound familiar?

If you work with cells in 3D…

Your experiment ends as one average per condition.
Cells are tracked by hand, or not at all.
Pore size comes from one flat image.
Reviewers ask for forces and mechanisms you cannot measure.

It doesn’t have to be this way. Meet MatriXense Studio.

Sound familiar?

If you develop drugs or cell therapies…

Your 3D assay ends in one endpoint readout.
Imaging data piles up faster than anyone can analyse it.
Hits from flat assays fail in human tissue.
You need human-relevant data before animal studies.

It doesn’t have to be this way. Meet MatriXense Studio.

MatriXense Studio

One gel.
Four chapters.

From the matrix to the cell, to how they shape each other. One app, every answer connected.

1 · The matrix3D FibreNetFibres, pores, alignment
2 · The cellLabel-free 3D TrackingTracks, migration, rhythms
3 · Cell–matrix3D Traction ForceForces, deformation, remodelling
4 · Cell–cellCell InteractionsPursuit, contact, killing

Chapter 1 of 4

The matrix.

Know your gel before the cells go in.

The matrixThe cellCell–matrixCell–cell

The matrix · 3D FibreNet

Know your gel before the cells go in.

Collagen · 30 × 30 × 30 µmDrag to turn

Pore size, seven ways

−60 %−30 %0+30 %

Error against the true 3D pore size, over 4 simulated networks with known truth.

Module · The matrix

3D FibreNet

Know the gel your cells live in, in true 3D.

Why it matters

Stiffness, pores and alignment steer how cells move, invade and activate. Check every gel before a cell goes in.

Questions it answers

  • Is my gel the same from batch to batch?
  • Does crosslinking change the pores cells squeeze through?
  • How aligned is the network around my cells?

8 analyses · add the ones you need

Fibre networkLength through crossings, junctions, connectivity
Fibre diameterCorrected for the microscope blur
Waviness and curvatureHow straight or crimped fibres are
Pore size, seven waysEvery common definition, against 3D truth
OrientationIn-plane alignment and out-of-plane order
3D from sparse stacks3D values with 80 % intervals
Typical image per conditionClosest to the median, not cherry-picked
Statistics and figuresTests, stars, Excel and Prism export

Chapter 2 of 4

The cell.

What each cell does on its own: where it is, how it moves, and its rhythm.

The matrixThe cellCell–matrixCell–cell

The cell · Label-free 3D Tracking

No stain. Just light.

Brightfield · z = 40 µm0 cells in focus
48/48

Cells found in brightfield, without a label (validation with known truth).

96%

Of the links the assistant is sure about were right. You only check the hard ones.

88%

Cell divisions called correctly, building full family trees.

The cell · migration

How cells move. In 3D, in numbers.

400 × 400 × 200 µm · 12 h0 h
PersistentRandomDrag to turn

Mean squared displacement

Speed and persistence in-plane by default, where brightfield z is less precise. Direction and depth in 3D.

The cell · spectroscopy

Go deeper
than averages.

Cell spectroscopy finds how each cell pulses, probes and changes shape. Rhythms you cannot see by eye.

Every cell type gets its own spectral barcode.

Fibroblasts

Macrophages

Simulated example. Wavelet power of cell shape over 6 hours; outlined = significant above noise.

Module · The cell

Label-free 3D Tracking

Every cell, found and followed in 3D. No stain, just brightfield.

Why it matters

Follow every cell for days without a dye that changes its behaviour. Works on primary and patient cells from day one.

Questions it answers

  • How fast and how directed do my cells move in 3D?
  • Which cells divide or die, and when?
  • Do subpopulations behave differently?

8 analyses · add the ones you need

Cells without a labelFound in brightfield z-stacks
Shape, volume, dry massFor every cell, at every timepoint
MigrationSpeed, persistence, MSD, direction
Family treesDivisions and deaths, full lineages
Tracking assistantChecks every link, asks only when unsure
Cell typesLabel, gate, cluster or teach a model
Cell spectroscopyRhythms and spectral barcodes
Treatment and eventsBefore and after, around any event

Chapter 3 of 4

Cell–matrix.

Cells pull. The matrix answers.

The matrixThe cellCell–matrixCell–cell

Cell–matrix · 3D Traction Force

See every pull. Measure every force.

Strain energy density0 min
0.63pJ

Strain energy one fibroblast puts into its matrix in 2 hours.

0.024µm

Bead localisation noise, so even small pulls count.

2–17%

Force error per cell when neighbours touch, against 47–232 % fitting each cell alone.

Cell–matrix · from beads to forces

From beads to forces.

Beads at two timepointsreference later
1 · BeadsEvery bead found in 3D at two timepoints.xref, x(t)
2 · Drift and matchingGlobal drift removed, beads matched with a calibrated 3D fit.x(t) − d(t) ↔ xref
3 · Displacement fieldInterpolated in 3D. Below the noise floor shows as noise.u(x) = x(t) − d(t) − xref
4 · Strain energyLinear elastic gel with your measured stiffness E.ε = ½(∇u + ∇uᵀ) · W = ½∫σ:ε dV

Cell–matrix · remodelling

Cells rebuild their matrix.

Two fibroblasts · collagen0 min
Fibre directionaligned to cell–cell axis

Densification by distance from the cell

Fibre alignment (order parameter S)

Densification at 0–25 µm matches a measured fibroblast (+5.1 %). Other values are a simulated example.

Module · Cell–matrix

3D Traction Force

How hard each cell pulls, and how it reshapes its matrix.

Why it matters

Contractility is a functional readout of cell state: myofibroblast activation, invasion and drug response, measured per cell in 3D.

Questions it answers

  • Do activated fibroblasts pull harder?
  • Does a drug relax contraction, and at what dose?
  • How do cells remodel the matrix around them?

8 analyses · add the ones you need

Beads in 3DFound and matched with a calibrated fit
Drift correctionStage and gel drift removed
Displacement fieldWith beads, or from the collagen itself
Strain energy and force mapsPer cell, in 3D, over time
Force qualityA measured noise floor on every map
Many cells at onceJoint fitting keeps each cell’s own force
Matrix remodellingDensification and fibre alignment
Gel movement in 3DThe whole gel, in three dimensions

Chapter 4 of 4

Cell–cell.

What cells do to each other.

The matrixThe cellCell–matrixCell–cell

Cell–cell · Cell Interactions

Watch cells interact. Then count it.

0

contacts

0

pursuing now

MacrophageCancer cellFibroblast

Module · Cell–cell

Cell Interactions

What cells do to each other, counted and tested.

Why it matters

Immune killing, tumour–stroma crosstalk and cell-therapy potency all happen between cells. Now you can put a number on them.

Questions it answers

  • Do my T cells or macrophages find and kill their targets?
  • Do cancer cells avoid immune cells?
  • Do cells prefer their own kind?

8 analyses · add the ones you need

Pursuit and attractionWho follows whom, and how directly
Contact and killingEvery contact, its duration and outcome
AvoidanceCells that keep their distance
Moving togetherAlignment and collective motion
Who meets whomSame or different type, partners over time
Cell–cell forcesThe matrix bridge between cells
Community behaviourHow a population organises itself
Is it real?Tested against randomised tracks

Bringing it together

Every cell,
a fingerprint.

Not one average per condition.

30+ measurements for every cell: force, motion, shape, matrix and contacts.

Cell states appear on their own. No marker needed.

One number per condition1 value

With MatriXense

Your next experiment.

MorningImage your gels.The way you always do. Any common microscope format.
EveningDrop the folder in.Plates, wells and timepoints are read from the file names.
OvernightEvery cell, measured.Found, tracked and measured. Every force mapped.
Next morningResults, ready.Figures, statistics and the methods text.

Work that used to take weeks, done while you sleep.

For your lab

From experiment
to paper.

Statistics, typical cells and a figure at journal size. In the same app, the same afternoon.

PNGSVGCSVExcelPrismMethods text

For your pipeline

From one gel
to a whole screen.

Whole plates, overnight, unattended. Every well analysed the same way, into one table.

Batch & Screening is in beta. Early partners shape it.

WellCompoundStrain energySpeedContacts

The 3Rs · translational

Fewer animals.
Better candidates.

Screen widely in 3D human-cell models first. Take only the strongest candidates further.

Patient cells ex vivoCell-therapy potencyMechanical biomarkersHuman-relevant preclinical data
1,000compounds tested in 3D human-cell models
40hits with a clear mechanical effect
5candidates taken into animal studies

Illustrative numbers. The 3Rs are written into EU Directive 2010/63/EU (Article 4).

Module · At scale

Batch & Screening Beta

Whole plates, overnight, analysed the same way.

Why it matters

The same pipeline for every well removes manual bias and makes large studies possible. Statistics use the well as the unit, not the cell.

Questions it answers

  • Which compounds change contractility or migration?
  • At what dose? What is the IC50?
  • Do donors or patients respond differently?

8 analyses · add the ones you need

Folders and platesPlate, well, z and t read from file names
Plate mapsDoses, controls and replicates per well
Queue overnightSurvives restarts, resumes where it stopped
One merged tableEvery well and every cell, one file
Plate QCZ′ per plate, problem wells flagged
Dose–responseIC50, Hill slope and hits per readout
Mechanical signatureEvery readout per compound, side by side
AtlasPer-cell fingerprint map across plates

Why MatriXense

One app, not a patchwork.

01Complete.Matrix, cells, forces and interactions in one app, with one set of results.
02Proven.Tested against known truth. Core methods peer-reviewed and published.
03Private.Runs on your own computer. Nothing uploaded. Works offline.

Accuracy

Measured. Not assumed.

0.024µm

Bead localisation noise on real time-lapse data.

48/48

Cells detected without a stain.

7

Pore-size definitions compared side by side.

88%

Cell divisions called correctly.

3×

Lower z-velocity error after 3D smoothing.

96%

Confident tracking links correct.

Acta Biomaterialia 2014Collagen network topology · Franke, Sapudom et al.
Scientific Reports 2017Label-free 3D cell tracking · Sapudom et al.
Biomaterials Science 20183D collagen matrix models · Sapudom & Pompe
Dr. Jiranuwat Sapudom

Dr. Jiranuwat Sapudom

Founder · Biochemist, Dr. rer. nat., Leipzig University

Our story

Born in the lab.
Built for every lab.

More than a decade of 3D collagen matrices, immune cells and mechanobiology. MatriXense exists because the analyses we needed did not.

2014Collagen network topology
2017Label-free 3D cell tracking
2026MatriXense Studio
NextDeep learning, fibre tracing, nonlinear forces

Privacy

Your data stays on your computer.

Nothing uploaded.

The engine runs locally and reads your images where they are.

Works offline.

A signed licence key locked to your computer. No account needed.

Fast on your hardware.

NVIDIA GPU workstations, Apple Silicon Macs, or CPU only.

LIFCZIND2OME-TIFFMicro-ManagerBioTek Gen5Operaand more

In one app

What you get.

✓Your gel, measured in 3D.Fibres, pores and alignment with error bars.
✓Every cell, tracked without a label.Shape, migration, divisions and rhythms.
✓Every pull, mapped to a force.Strain energy, deformation and remodelling.
✓Every interaction, counted and tested.Pursuit, contact, avoidance and killing.

One app. On your own computer. Your data never leaves it.

Getting started

Start small. Grow with your study.

1
Try it

Every module, on your own images, with an onboarding call.

2
Pilot

Or send us one dataset. We analyse it, and the fee is credited to a licence.

3
Licence

A yearly licence quoted for exactly the modules and analyses you need.

4
Support

All updates, new methods and help from scientists, for as long as it runs.

Core facility & site licences Methods text for your paper Study and screening analysis Custom software and integration NDA and data processing agreement

MatriXense Studio

See your cells in 3D.
In numbers.

Get in touch

hello@matrixense.com

matrixense.com

Animations use simulated data. MatriXense software is for research use only.

1 / 14
Turn your phone sideways for the full presentation.