The Hidden Trap in Your Western Blot (Why Your Analysis Might Lie)
“We saw a reduction in Protein X expression.” Are you sure?
We've already covered how to avoid Western blot failures and statistical analysis mistakes.
But there's still a small, hidden danger that can hurt your data.
A 2016 BMC Biology article titled “Quantifying Western Blots: None More Black” revealed a brutal truth: Most quantification done on Western blots is built on shaky ground.
Here’s why.
The Tubulin Trap
In one example, researchers used tubulin as a loading control. Everything looked fine—equal bands, clean bar chart, p < 0.05.
But here’s what really happened:
They accidentally loaded more protein in the wild-type lane. The chemiluminescent film was saturated, meaning the “equal” tubulin bands were a visual illusion.
The quant software couldn’t detect the difference, and the conclusion? Totally wrong.
Why This Happens All the Time
Quantification software doesn’t know when a signal is maxed out.
Why? Because this might happen before pixels turn red.
Once you leave the linear range, everything looks the same on screen—even if you doubled the protein.
Bell’s study showed that at higher concentrations, multiple lanes gave the same signal intensity reading, even though the amount of protein was increasing.
So, when you base conclusions on grayscale bar heights? You might be reading shadows, not science.
Housekeeping Genes: Helpful or Harmful?
Proteins like β-actin or tubulin are popular for normalization.
But they:
Are often overexpressed relative to your target
Get saturated faster than your protein of interest
May vary between tissues and treatments
This means your “stable control” might be silently shifting beneath your data.
When Exposure Goes Wrong
In another example, researchers overexposed some lanes to visualize faint bands.
The result? They blew out the signal in other lanes—making relative quantification impossible.
You can’t analyze grayscale when some parts are pure white and others are nearly invisible.
So What Can You Do Instead?
Better practices (straight from the paper + Wildtype One's Playbook):
Check your exposure is within the linear range—run a dilution series to test it
Quantify at two different levels of exposure and compare
Use multiple housekeeping proteins or switch to total protein staining (e.g., stain-free gels)
Validate key findings with orthogonal methods like ELISA or Wes (Simple Western)
Replicate blots with both biological and technical replicates
Avoid reprobing stripped membranes—they almost always give distorted signals
Final Thought
Western blots don’t lie. But they do mislead—quietly, consistently, and with the full confidence of a bar chart.
Don’t be that lab that trusted pixel density without checking exposure.
Much love,
Carl from Wildtype One