Research vocabulary is most useful when it prevents a specific misunderstanding. The following terms help separate a molecule's chemistry, the quality of a material and the strength of a clinical claim.

A peptide is a chain of amino acids connected by peptide bonds. Sequence means the ordered amino acids; a fragment is part of a longer chain. An analogue is a modified relative, not necessarily an interchangeable product. NHGRI provides the underlying peptide definition. NHGRI: Peptide definition

Identity asks what substance is present. Assay measures an amount or activity under a defined method. Purity describes the proportion measured as the intended material relative to specified alternatives. An impurity is an unintended component; a degradation product results from change over time. A certificate of analysis records selected tests for a stated sample or lot, not every possible property.

In vitro means an experiment outside a living organism, such as in cells. Ex vivo commonly describes removed tissues studied outside the body. In vivo means within a living organism and does not necessarily mean human. Preclinical evidence precedes or supports clinical development; it should retain its model label.

A control provides a comparison. Randomization assigns groups by chance. Blinding limits knowledge of assignment that might influence care or measurement. An endpoint is a defined outcome; a primary endpoint is the main outcome selected for the principal test. NIH's clinical-research glossary explains these study concepts. NIH: Clinical research glossary

A biomarker measures a biological feature. A clinical outcome concerns how people feel, function or survive. Statistical significance does not by itself establish an important benefit. Investigational describes a product being studied; approved refers to a specific regulatory authorization.

The habit behind this glossary is simple: whenever a claim sounds persuasive, ask which term is doing the work. Human cells are not human participants, an assay is not approval, and a proposed mechanism is not a demonstrated treatment outcome.