How to reproduce a Western blot from a paper

By the Vara team · Updated 2026-10-07

This page explains how to rebuild a published Western blot so that your bands can be compared with the figure. A typical methods section names the antibody and a dilution such as 1:1000, then stops before lysis, loading, transfer, and exposure details that decide what the blot looks like. Below are the gaps to fill, the parameters that move together, and a triage table for the most common failures.

What methods sections usually leave out

Antibody clone, host species, and lot or catalog identifier
Two antibodies raised against the same protein can recognize different epitopes and give different band patterns. A dilution ratio is meaningless without knowing which antibody it applies to.
Amount of total protein loaded per lane
Signal depends on how much protein is on the membrane. If the authors loaded 40 µg and you load 10 µg, a weak band may simply be below detection.
Lysis buffer and sample preparation
Detergent strength, protease and phosphatase inhibitors, and whether samples were boiled all affect membrane proteins, phospho-epitopes, and aggregation.
Transfer method, membrane type, and transfer time or current
Wet versus semi-dry transfer, PVDF versus nitrocellulose, and transfer duration change how efficiently small and large proteins reach the membrane.
Blocking agent and the buffer used for antibody dilution
Nonfat milk contains phosphoproteins and biotin, so it can raise background with some phospho-specific antibodies or streptavidin detection. BSA is often preferred in those cases.
Normalization strategy and exposure
Papers often show a loading control without saying whether quantification was normalized to it or to total protein staining, or whether the exposure was within the linear range.

Key parameters and what they change

Parameter	Typical	Changes together with
Primary antibody dilution	Often 1:500 to 1:5000, set by the antibody's datasheet and titration	Antibody volume per blot, incubation time and temperature, and exposure time. A more concentrated antibody usually needs a shorter exposure and can raise background.
Primary incubation	Commonly overnight at 4 °C or 1 to 2 hours at room temperature	Dilution. Shorter, warmer incubations often need a more concentrated antibody to reach similar signal.
Protein load per lane	Often 10 to 50 µg of total lysate	Lysate concentration from the protein assay, volume per lane, and well capacity of the gel.
Gel percentage	Chosen by target size; lower percentage for large proteins, higher for small ones	Separation of nearby bands and transfer efficiency for the target size.
Incubation volume	Enough buffer to keep the membrane fully covered, scaled to membrane area	Total antibody needed. Change the membrane size or number of blots and the antibody amount changes with it.

Troubleshooting

Symptom	Likely causes	What to check first
No band at the expected size	Target not expressed in your sample, too little protein loaded, poor transfer, or an antibody that does not work in your species or conditions.	Stain the membrane with Ponceau S to confirm transfer, include a positive control lysate, and confirm the antibody is validated for Western blot in your species.
High, even background across the membrane	Blocking too short, antibody too concentrated, or insufficient washing. Patchy rather than even background usually means the membrane dried out.	Extend blocking, dilute the primary and secondary antibodies further, add washes with detergent, and keep the membrane wet at all steps.
Extra bands at unexpected sizes	Cross-reactivity, protein degradation, post-translational modification, splice isoforms, or secondary antibody binding on its own.	Run a secondary-only control, add fresh protease inhibitors, and compare with a knockdown or knockout sample if one is available.
Smeared or distorted bands	Overloaded lanes, high salt or lipid in the sample, incomplete denaturation, or running the gel too hot.	Load less protein, clarify lysates by centrifugation, and run at lower voltage or with cooling.
Band intensity does not track with the treatment shown in the paper	Signal saturated during exposure, or the loading control itself changes with treatment.	Take several exposures and confirm the signal is in the linear range; consider total protein normalization.

Worked example: from a methods sentence to executable steps

Cell lysates were separated by SDS-PAGE and transferred to PVDF membranes. Membranes were blocked and incubated with anti-target antibody (1:1000) overnight at 4 °C, followed by HRP-conjugated secondary antibody, and visualized by ECL.

Each step is tagged by how clearly the text states it: explicit, partial, inferred, or missing.

1.
Lyse cells and measure protein concentrationPartial
Lysis is stated; the buffer, inhibitors, and protein assay are not.
2.
Load equal protein per lane and run SDS-PAGEPartial
Amount per lane and gel percentage are not given.
3.
Transfer to PVDF membraneExplicit
4.
Transfer method and durationMissing
Wet or semi-dry, current, and time are needed to match transfer of the target size.
5.
Block the membranePartial
Blocking agent and time not stated; milk versus BSA matters for some antibodies.
6.
Incubate with primary antibody at 1:1000, overnight at 4 °CExplicit
The antibody source and clone still need to be found in the reagent table.
7.
Incubate with HRP-conjugated secondary antibodyPartial
Stated; dilution and time are not given. The datasheet range is a starting point, not the authors' value.
8.
Detect by ECL and imagePartial
Exposure settings and whether quantification used a loading control are not stated.

How Vara helps

Upload the PDF or a photo of the methods, and Vara splits the blot into steps with parameters, tagging each one explicit, partial, inferred, or missing.
Gaps like protein load, transfer conditions, and blocking agent appear as listed reproduction risks rather than quietly filled defaults.
Antibody dilution, incubation volume, and number of blots are linked by formulas, so changing the membrane count recalculates how much antibody to prepare.
Antibodies and buffers can be stored in the materials library, and each run is logged against the protocol version so you can see which dilution produced which blot.

Try Vara on your own protocol See how importing works

Free calculators for this protocol

FAQ

Should I block with milk or BSA?+

Nonfat milk is inexpensive and works for many antibodies. BSA is often preferred for phospho-specific antibodies, because milk contains phosphoproteins, and for biotin-streptavidin detection, because milk contains biotin. Follow the antibody datasheet when it specifies a blocking agent.

Why does my Western blot show no bands when the paper shows a clear one?+

Start by checking transfer with a reversible total protein stain, then confirm the target is expressed in your cells with a positive control. Common causes are a lower protein load, a different antibody clone or lot, or transfer conditions that do not suit the target size.

If I change the primary antibody from 1:1000 to 1:500, what else should I change?+

A higher antibody concentration usually gives more signal and more background. You may need a shorter exposure, more thorough washing, or a shorter incubation. Change one variable at a time and record it so the comparison stays interpretable.

Is a single loading control enough for quantification?+

It can be, if the loading control is stable under your conditions and both signals are within the linear range of detection. Many journals and reviewers now favor total protein normalization because common housekeeping proteins can change with treatment.