Scratch Assay Imaging Protocol: A Step-by-Step SOP for Analysis-Ready Wound Images

This protocol produces scratch assay images that segmentation software can actually measure: a matched image set where every field is identically framed, focused, lit, and exposed at every time point, ready for wound area and migration rate analysis.

The one rule this scratch assay imaging protocol exists to enforce: your software compares pixels, not biology. Anything that changes between Time 0 and Time 24 other than your cells will be measured as migration. Every step below exists to hold something constant.

It covers transmitted-light (phase contrast) imaging of 2D scratch/wound healing assays in multi-well plates or dishes, on any inverted microscope, and it covers acquisition only. For the assay itself, see our wound healing assay protocol; for the reasoning behind each rule, see our scratch assay imaging guide.

Analyze your scratch assay image!

Equipment and settings for scratch assay imaging

Item	Requirement
Microscope	Inverted, phase contrast
Objective	4x or 10x (both wound edges must fit in one frame)
Objective NA	<0.5 NA for standard plastic plates (must focus through thick plastic)
Camera	Any; must allow manual exposure
Stage	Heated 37 °C stage if lid condensation blurs the field
Calibration	Stage micrometer
File format	TIFF or PNG (never JPEG)
Target resolution	200–500 pixels across the wound width

Part 1: Set up the microscope (once per project)

1. Calibrate your pixel size with a stage micrometer at the objective you will use. Do this before any experiment or your measured areas will not correspond to real dimensions.

2. Select a 4x or 10x objective and confirm both wound edges sit inside one frame. A ~500 μm gap fits comfortably at 10x on a standard camera.

3. Check the pixel count across the gap. Adjust magnification until the wound spans 200–500 pixels. Below this range, thresholding renders the boundary jagged and imprecise.

4. Fully open the field and condenser diaphragms.

5. Focus on the sample, then do not touch the focus knobs again until step 8.

6. Close the field diaphragm to its smallest diameter.

7. Raise or lower the condenser until the edges of the field diaphragm appear sharp, overlaid on your sample.

8. Center the field diaphragm with the condenser's adjustment knobs.

9. Open the field diaphragm just to the edge of the field of view. Opening it further adds scatter and glare; leaving it too closed creates a hard dark circle in your image.

10. Verify phase ring alignment. Confirm the objective's phase ring (PH1, PH2, etc.) is concentric with the matching condenser annulus, using a Bertrand lens or the empty ocular position. Mismatched rings are a leading cause of uneven shading.

11. Record every setting: objective, magnification, camera adapter, sensor, binning, gain, illumination intensity, condenser position. You must reproduce all of them at every later time point. A magnification error of only 3–5% meaningfully distorts wound-width measurements.

Part 2: Capture Time 0

12. Start within 15 minutes of making and washing the scratch, before debris settles or migration begins in earnest.

13. Turn off auto exposure and set exposure manually. Auto exposure re-balances between a mostly-empty Time 0 frame and a mostly-cell Time 24 frame, making identical cells look different to your software.

14. Set exposure so both the cells and the wound edge retain texture. Do not brighten a flat image with exposure or gain; fix flatness by re-aligning illumination and adjusting the condenser instead. Clipped white areas and crushed dark areas are both unrecoverable for thresholding.

15. Position the scratch horizontally and centered in the field of view.

16. Skip the first and last ~5% of the scratch length. Refraction at the well edge overexposes those regions and makes them unusable.

17. Capture images covering ~90% of the scratch length, with no overlap between frames.

18. Record the stage coordinates and focus Z-position for every field. You will return to these exact positions; "finding the wound again" by eye is one of the largest hidden sources of error in this assay.

19. Save as TIFF or PNG, one folder per scratch. JPEG compression introduces block artifacts and altered intensities that corrupt thresholding.

20. Return the plate to the incubator immediately.

Part 3: Capture later time points

21. Restore every recorded setting from step 11 before imaging. Change nothing.

22. Return to the recorded stage coordinates, not to a similar-looking region. A scratch is never uniform along its length, so a different stretch of the same wound reads as migration.

23. Return to the recorded Z-position. Use autofocus if available. Never refocus by eye "until it looks good."

24. Confirm exposure is still locked at the Time 0 value.

25. Capture the same fields, in the same order, in the same file format.

26. Stop when the gap is roughly half closed. There is usually no need to image to full closure, and longer incubation lets proliferation confound your migration measurement.

Acceptance criteria: is your image set analysis-ready?

Your image set is analysis-ready when all of these are true:

•  Both wound edges are visible in every frame.

•  The wound spans 200–500 pixels.

•  Illumination is even, with no dark circular zone.

•  Cells and wound edge both retain texture; nothing is clipped white or crushed black.

•  Every time point uses identical optics, exposure, gain, and focus.

•  Every time point images the identical field, by recorded coordinates.

•  All files are TIFF or PNG.

•  Pixel size is calibrated for the objective used.

If any box fails, fix it and re-image rather than analyzing. No software recovers a broken acquisition.

Troubleshooting your scratch assay imaging protocol

Symptom	Cause	Fix
Dark circular shadow, center of frame	Condenser misaligned, objective not seated, field diaphragm too closed, or vignetting	Re-run steps 4–10; reseat objective; image a blank dish to confirm it is optical, not sample
Wound width changes with no visible cell movement	Focus or working-distance drift	Return to the recorded Z-position; use autofocus
Blurry wound edges	Wrong focal plane, plate not flat, NA too low	Refocus on cell edges (not the empty gap); check plate flatness
Software finds several small holes, not one gap	Debris in the wound	Avoid debris-prone fields; enable "include holes" in ImageJ's Analyze > Particles
Washed-out or flat images	Auto exposure on, or poor Köhler alignment	Lock exposure (step 13); re-run steps 4–10
Overexposed frame edges	Refraction at the well periphery	Skip the outer ~5% of scratch length (step 16)
Wound in a different position at T24	Field-of-view drift	Image by recorded stage coordinates (step 18); use fiducial marks
Jagged boundary under thresholding	Too few pixels across the gap	Increase magnification to reach 200–500 px (step 3)
Thresholding inconsistent across a plate	JPEG artifacts	Re-save as TIFF or PNG; re-image if originals were JPEG

Notes

Upstream: A uniform, reproducible scratch removes one whole axis of variability before imaging starts, which is what a tool like CytCut is for. Consistent wound width and position make relocating the same field far easier.

Downstream: Once acquisition is clean, hand the folder to an automated analyzer such as Sophie's Scratch Analyzer for wound area, closure percentage, and migration rate. Good imaging is the prerequisite; no analyzer rescues a broken image set.

Need the full protocol for a different biomedical assay?

FAQ

• Can I use brightfield instead of phase contrast? Not reliably. Brightfield rarely gives enough contrast to separate a thin unlabeled monolayer from the empty gap. Phase contrast enhances cell borders without labels and is available on nearly every cell-culture microscope.

• How do I know the dark circle is optical and not biological? Image a blank dish. If the dark zone persists, it is your optics (re-run steps 4–10). If it disappears, check your sample: meniscus, debris, or lid condensation.

• Do I really need Köhler illumination? Yes, if your microscope allows the adjustment. It is the single highest-return step in this protocol, and it directly fixes the uneven lighting and glare that break automated segmentation. Some cell-culture scopes are preset and cannot be adjusted.

• How often should I image? Sample often enough to fit a migration rate, and stop when the gap is about half closed. Imaging to full closure is usually unnecessary and increases the risk that proliferation confounds migration.

• What if I already saved everything as JPEG? Re-image if you can. JPEG artifacts alter pixel intensities, and re-saving a JPEG as TIFF does not recover the lost information.

References

• Jonkman, J.E.N. et al. (2014). An introduction to the wound healing assay using live-cell microscopy. Cell Adhesion & Migration, 8(5):440–451.

• Jonkman et al. (2014) — full text (PMC5154238).

• Kauanova, S., Urazbayev, A. & Vorobjev, I. (2021). The Frequent Sampling of Wound Scratch Assay Reveals the "Opportunity" Window for Quantitative Evaluation of Cell Motility-Impeding Drugs. Frontiers in Cell and Developmental Biology.

• Imaging and Analyzing Your Wound Healing Assay — Bitesize Bio.

• Wound Healing and Migration Assay: Principle — ibidi.

• CLYTE — How to do Wound Healing (Scratch) Assay for Cell Migration Analysis.