01Identify everything that differs between conditions

Verheijen and colleagues’ original study found molecular changes in cardiac and hepatic microtissues exposed to DMSO, a commonly used laboratory solvent. The finding supports checking vehicle effects in the relevant model rather than assuming that a familiar solvent is biologically inert.Verheijen and colleagues — DMSO induces drastic changes in human cellular processes and epigenetic landscape in vitro (opens in a new tab)

Read the final mixture experienced by the cells or assay components, not just the stock preparation. The name of the stock solvent does not tell you its final proportion in every condition.

Record the vehicle composition alongside the peptide concentration. If other additives accompany the peptide, keep them in that comparison rather than treating them as invisible background.

The vehicle may have more than one component. A control matching only the most obvious solvent can leave other differences between the reference and test mixtures unresolved.

This is a documentary check: identify what the study actually compared before deciding which component deserves credit for a change.

02Distinguish untreated and vehicle-matched comparisons

The fictional peptide mixture is 40 units below untreated, but 20 units below the matched vehicle. Those are different comparisons, and a statement assigning all 40 units to the peptide would overlook the vehicle-associated difference.

The percentage basis also matters: the decrease from 100 to 60 is 40%, while the decrease from 80 to 60 is 25%. Neither percentage should appear without its reference.

This does not prove that the effects of solvent and peptide add independently. The observations describe the tested combinations; they do not establish the peptide’s hypothetical effect in a solvent-free system.

When a paper normalises the vehicle response to 100%, read the unnormalised context if available. Normalisation can make the peptide comparison clear while hiding whether the vehicle itself altered the untreated baseline.

03Check whether the vehicle changes across the series

In a concentration-response experiment, ask whether increasing peptide concentration also changes the final vehicle amount. If both vary together, the shape of the curve may not isolate the peptide’s contribution.

A single vehicle-control condition cannot represent several different vehicle mixtures without a justification. Read how the authors accounted for the actual conditions across the series.

For example, if the largest peptide addition also introduces the largest solvent fraction, a response confined to that condition needs both changes considered. The pattern alone cannot tell you which change caused it.

Time is part of that exposure. A vehicle introduced at a different stage or retained for a different interval can make the nominally matched comparison less direct.

04State what the matched comparison establishes

If the vehicle-matched comparison is well characterised, the result can describe the effect associated with adding the peptide in that formulation. That is a useful and specific conclusion.

It does not establish that the same response would occur in every alternative formulation. A change of vehicle changes the experimental context and needs its own evidence.

Also keep the measurement target visible. A vehicle may leave one endpoint similar while changing another; lack of a visible survival effect is not universal evidence of biological neutrality.

A concise summary names the final vehicle, reference condition and measured endpoint. If the mixture is incompletely reported, retain that uncertainty instead of inventing a matched control. This allows the reader to assess the peptide-associated result without assigning unexplained formulation effects to the peptide.