Quantify western blot bands online

Nothing to install · No sign-up · Your images never leave your device

Load a blot, click each band, and get background-corrected intensities normalised to your housekeeping protein. Everything runs inside your browser — no image is ever sent to a server. Results open in a new tab, ready to export as CSV or print for your lab notebook.

Works in Chrome, Edge, Firefox and Safari. For the sharpest measurements, use the file your imager produced rather than a screenshot.

How it works

1. Load your housekeeping image into the first slot, then load your target image into the second. If both proteins are on the same image, load the same file into both slots. Press + Add target for up to three targets sharing one housekeeping control.
2. Click each band. A box appears and centres itself on the band. Drag the corners to resize it, drag the box to move it, and use the handle below the box to rotate if your scan is slightly tilted. Lanes are numbered left to right, so the order you click in doesn't matter.
3. Press Run. Results open in a new tab: raw intensity, the target-to-housekeeping ratio, and the relative value for every lane.
4. Tick your control lanes. Relative values recalculate instantly. Tick more than one and their mean becomes the baseline. Export the table as CSV, or print it as a PDF.

If you crop away empty margins first, nothing is lost — cropping copies pixels one to one, so no value is altered, and any boxes you already drew are kept.

The calculation

Every box is measured as a background-corrected integrated density:

net = Σ(signal) − median(surrounding ring) × area

• Signal is 255 − brightness for dark bands on a light background, and brightness for bright bands on a dark background. Polarity is detected from each image automatically, so there is nothing to set — a film scan and a fluorescence image can be used together.
• Background is the median of a ring of pixels immediately around the box, multiplied by the box area. Using the median rather than the mean keeps a neighbouring band from inflating the estimate.
• Ratio = target net ÷ housekeeping net for the same lane.
• Relative = that lane's ratio ÷ the mean ratio of the lanes you ticked as controls.

Because background is removed per unit area, boxes do not have to be the same size. They only need to contain the whole band with a little room to spare. A box that clips the edge of a band is the one thing that does bias the result, so an edge of the box turns red where it cuts through signal.

Rotated boxes read the original pixels directly using nearest-neighbour sampling. The image itself is never rotated, so no interpolation changes a measured value.

Why normalise to a housekeeping protein?

Raw band intensity reflects more than how much protein was present. Small differences in how much lysate was loaded, how evenly it transferred, and how long the blot was exposed all change the number. Dividing by a housekeeping protein from the same lane cancels most of that, which is why relative values — not raw intensities — are what gets reported.

Two conditions have to hold for the correction to mean anything. The housekeeping protein must come from the same membrane and the same loading as the target, and its own signal must be strong but not overexposed. If your control lanes disagree with each other by more than about 30%, the results page flags it — that usually points to a loading or transfer problem rather than a real biological difference.

Frequently asked questions

Are my images uploaded anywhere?

No. The tool is a single file that runs in your browser, and there is no server component to send anything to. You can verify this yourself: open your browser's developer tools, switch to the Network tab, and analyse a blot. No request carries your image.

Can I use a JPEG?

You can, but you shouldn't for anything you plan to publish. JPEG is a lossy format, so it changes pixel values — and pixel values are exactly what densitometry measures. Use PNG, or whatever lossless format your imager exports.

Does it work with 16-bit TIFF files?

Not yet. Browsers cannot decode TIFF natively, so a 16-bit TIFF has to be exported as PNG first. Your imaging software can do this. Note that converting to 8-bit reduces the number of intensity levels from 65,536 to 256, so export at the widest range your software offers.

Can I quantify more than one target protein?

Yes — up to three targets, all normalised to the same housekeeping control. Each target gets its own image slot and its own rows in the results table.

What if my housekeeping protein is on a different membrane?

Then it cannot normalise that target. Loading and transfer differences are specific to a membrane, so the correction only works when the housekeeping signal comes from the same membrane as the target — whether by stripping and reprobing, cutting the membrane, or multiplex detection.

How is the background subtracted?

Locally, per band. The tool takes the median intensity of a ring of pixels surrounding each box and subtracts it in proportion to the box area, so a blotchy or gradually shaded membrane affects each lane on its own terms rather than being corrected globally.

Can I save an analysis and come back to it?

Not currently. Because nothing is stored — not on a server and not in your browser — closing the tab ends the session. Export the CSV before you leave.

How to cite this tool

If you use these values in a publication, you can describe the method like this:

Band intensities were quantified with Blot Reader (https://pymtools.castingfilter.com/western-blot-quantification, accessed DD Month YYYY) as background-corrected integrated density, with local background estimated from the median of a ring surrounding each region of interest, and normalised to [housekeeping protein] in the same lane.

Reviewers may ask how the numbers were produced, so it helps to keep the raw blot images and note which lanes you used as controls.

Related tools

• ΔΔCt calculator — relative qPCR quantification (coming soon)
• Dilution calculator — C₁V₁ = C₂V₂ (coming soon)

For research use only. Not for use in diagnostic procedures. Results are provided without warranty — verify them before publication.