AutoBlot Studio
Normalisation guide

Loading Control vs Total Protein Normalisation

GAPDH, β-actin, or a Ponceau S stain? Here's how to choose a normalisation method that will hold up under reviewer scrutiny — and how to tell if the one you're already using is reliable.

8 min read · Updated June 2026
1

Why normalisation matters

No two lanes on a western blot ever receive exactly the same amount of protein. Pipetting error, transfer efficiency, and membrane edge effects all introduce lane-to-lane variation that has nothing to do with your biology.

Normalisation divides your target band's signal by a reference signal measured in the same lane, cancelling out that loading variation so the remaining differences reflect your experimental treatment.

Key principle
A good normaliser is one thing only: constant across your experimental conditions. If it changes with treatment, it's no longer measuring "how much protein did I load" — it's measuring your treatment effect too, and your normalised data becomes meaningless.

2

Housekeeping protein normalisation

This is the traditional approach: probe the same membrane (often after stripping, or on a separate blot run in parallel) for a protein assumed to be expressed at a constant level — most commonly GAPDH, β-actin, α-tubulin, or vinculin.

Housekeeping normalisation Normalised value = Corrected target intensity ÷ Corrected housekeeping intensity (same lane)

Advantages: familiar to reviewers, requires no extra staining step beyond an additional antibody, works well with chemiluminescent detection on film or single-channel imagers.

Disadvantages: the central assumption — constant expression — is frequently false. GAPDH is upregulated under hypoxia and many metabolic interventions. β-actin changes with cytoskeletal remodelling, cell density, and differentiation state. Using a housekeeping protein that varies with your treatment doesn't just add noise — it can flip your result.

A real failure mode
If your treatment increases GAPDH expression and you normalise to GAPDH, a true increase in your target protein can be masked or even appear as a decrease. This is one of the most common — and hardest to spot — sources of irreproducible western blot results.

3

Total protein normalisation

Instead of relying on one protein as a proxy for "how much did I load", total protein normalisation stains every protein on the membrane and uses the summed signal in each lane as the denominator.

Total protein normalisation Normalised value = Corrected target intensity ÷ Total protein lane signal (same lane)

Common methods include reversible stains applied before antibody probing — Ponceau S (cheap, visible without an imager, but less sensitive) and fluorescent stains such as REVERT or No-Stain (more sensitive, compatible with fluorescent multiplex imaging, and don't require stripping).

Advantages: doesn't rely on the expression of any single gene staying constant, so it's far more robust to unexpected treatment effects. Stains applied before probing don't consume membrane real estate the way a second antibody does.

Disadvantages: requires a compatible imaging system (fluorescent stains need a fluorescent imager), and total signal can still be affected by transfer efficiency differences for very high or very low molecular weight proteins relative to your target.

Tip
If you're imaging on a fluorescent multiplex system, total protein stains are usually the easiest upgrade you can make to your workflow — one extra staining step before blocking, with no extra antibody cost.

4

Side-by-side comparison

Factor Housekeeping protein Total protein stain
Robust to treatment effects Only if validated Generally yes
Extra antibody cost Yes No
Requires membrane stripping Often No
Works on chemiluminescent imagers Yes Limited (Ponceau S only)
Reviewer familiarity Very high Increasing, now widely accepted
Current journal guidance Acceptable with validation Often preferred or required
Best of both
Many labs report both — a familiar housekeeping protein alongside a total protein stain. If the two normalisations agree, it strengthens confidence in the result; if they disagree, it's a signal to investigate which one is actually stable in this experiment.

5

How to validate your housekeeping protein

If you want to keep using a housekeeping protein, validate it under your actual experimental conditions before you rely on it:

Step What to do
1 Run your full experimental conditions (control + all treatments) on one blot.
2 Probe for your candidate housekeeping protein and quantify its raw, background-corrected intensity in each lane — without normalising.
3 Compare across conditions. If the intensity is statistically indistinguishable between groups, it's a valid normaliser for this experiment.
4 If it varies significantly, try an alternative (e.g. switch from GAPDH to vinculin, or to a total protein stain) and repeat.
Validate per experiment, not once
A housekeeping protein that's stable for one treatment (e.g. a drug) may not be stable for another (e.g. a hypoxia or starvation protocol). Re-validate whenever you change the experimental paradigm.

6

What journals expect in 2026

Reviewer expectations have shifted significantly over the past decade. Most journals in cell and molecular biology now either require total protein normalisation, or require justification and validation if a housekeeping protein is used instead.

What to include in your methods section regardless of which method you use:

Detail Why reviewers ask for it
Which normaliser was used and why Shows the choice was deliberate, not just "what the lab always does"
Validation data (if housekeeping protein) Demonstrates the normaliser doesn't change with treatment
Whether normaliser was probed on the same membrane Affects whether stripping/re-probing artefacts are a concern
Quantification software and version Reproducibility — lets others repeat your analysis pipeline

7

How to switch without redoing your blots

If you've already run your blots with a housekeeping protein but want to add total protein normalisation, you may not need to start over:

  • If you saved the membrane (rather than discarding after stripping), it can often still be stained with Ponceau S even after multiple probing rounds.
  • For future blots, stain for total protein before blocking and probing — it adds only a few minutes and doesn't interfere with downstream antibody binding.
  • Re-image any saved total protein stains and re-run your quantification with both normalisers to compare.
Switching normalisation method in AutoBlot Studio
AutoBlot Studio lets you toggle between housekeeping protein and total protein normalisation for the same dataset without re-detecting bands — useful for exactly this kind of comparison.

Frequently asked questions

Is total protein normalisation better than housekeeping proteins?

Most journals now consider total protein normalisation the gold standard because housekeeping proteins like GAPDH and β-actin can change expression under many experimental conditions. However, housekeeping proteins remain acceptable if you validate that their expression is stable under your specific conditions.

Can I use both a housekeeping protein and total protein stain?

Yes. Many labs run both — a housekeeping protein for a familiar reference and total protein staining (e.g. Ponceau S or REVERT) as the primary normaliser. Reporting both strengthens your figure.

What if my housekeeping protein changes between conditions?

If your housekeeping protein's expression changes with treatment, it is not a valid normaliser for that experiment. Switch to total protein normalisation or test multiple housekeeping candidates to find one that is stable under your conditions.

Do I need to re-probe the membrane for total protein staining?

No. Total protein stains such as Ponceau S or REVERT are applied before antibody probing and do not require stripping or re-probing, making them faster and gentler on the membrane than a second housekeeping antibody.

Normalise both ways in one click

AutoBlot Studio quantifies your bands once, then lets you switch between housekeeping and total protein normalisation instantly — so you can compare and report whichever is most defensible.

Try AutoBlot free →