01Identify exactly what the comparison changes

Semenova and colleagues explicitly identify the four-residue tuftsin chain and the seven-residue Selank chain in their comparative report. Reading the sequences side by side reveals the shared beginning and the additional Pro-Gly-Pro segment.Semenova and colleagues — Selank and tuftsin after PCPA pretreatment (opens in a new tab)

This alignment identifies a structural difference. It does not establish whether the extension changes target binding, degradation, distribution or several properties together. Those are competing explanations to investigate.

In a literature search, retain both exact names. A paper retrieved because it discusses tuftsin in its introduction may never have tested Selank, and a Selank experiment may not include a tuftsin control.

02Read the serotonin experiment as a defined comparison

The 2009 report studied Wistar rats pretreated with PCPA, an inhibitor of serotonin synthesis. At the reported early observation, Selank increased brain-stem serotonin metabolism; tuftsin did not produce that change and decreased the neocortical measure.Semenova and colleagues — Selank and tuftsin after PCPA pretreatment (opens in a new tab)

The pretreatment is part of the model, not incidental background. The result concerns a deliberately altered system. It should not be presented as a measurement of normal human mood or as evidence that more serotonin is always beneficial.

The regional distinction matters too. A finding in one sampled brain region cannot be expanded into a claim about the entire brain without measurements that support that expansion.

This article uses the report’s accessible English abstract. It does not infer unreported effect sizes, allocation procedures or complete adverse-event findings from the authors’ broader conclusion.

03Do not merge chemistry and behaviour into one endpoint

Kozlovskaya and colleagues’ 2003 study compared tuftsin-family peptides and Selank in rodent stress-related behavioural experiments. Its abstract describes several rat and mouse groups and sorting by emotional reactivity, with differences associated with the tested structures or their fragments.Kozlovskaya and colleagues — Tuftsin-family peptides in rodent stress experiments (opens in a new tab)

That report offers a different kind of observation from the serotonin experiment. A behavioural score is not a direct measurement of the chemical pathway responsible for it.

Even if two experiments point in a compatible direction, compatibility is weaker than showing that one measured change caused the other. A proposed mechanism needs a design that can distinguish it from alternative explanations.

04Write the question before ranking the materials

A useful comparison asks whether the extension changes a specified readout under matched conditions. “Which is stronger?” leaves the readout, exposure and comparison rule undefined.

For a hypothetical evidence table, place sequence confirmation in one row, intact-material persistence in another and the selected biological response in a third. An empty row stays empty even when another row contains a positive finding.

This also makes disagreement informative. If two materials share one response but differ in another, the evidence may be separating mechanisms rather than identifying an overall winner.