Research glossary.
Definitions to help you read the library and its original sources.
-
Adverse event
- An unfavorable occurrence during a study; its occurrence alone does not prove the intervention caused it.
-
Agonist
- A substance that activates a receptor. Activation alone does not establish a useful outcome in a whole organism.
-
Blend
- A combination presented together as one preparation. Evidence for separate ingredients does not automatically establish evidence for the combination.
-
Blinding
- Keeping relevant people unaware of group assignment to reduce bias.
-
COA
- Certificate of analysis: a report of specified tests and results for an identified sample.
-
Comparator
- The group or condition against which an intervention is assessed.
-
Concentration
- An amount of substance expressed per volume or other specified quantity.
-
Confidence interval
- A range expressing statistical uncertainty around an estimate under the analysis assumptions.
-
Endpoint
- A defined outcome measured in a study.
-
In vitro
- Research performed outside a living organism, often in cells or isolated systems.
-
In vivo
- Research performed in a living organism.
-
Lot / batch
- An identifier used to trace a particular production group of material.
-
Mechanism
- An explanation of how an effect may occur.
-
Peptide
- A molecule made from linked amino acids. Catalogs may group other compounds with peptides; the category name does not change a molecule’s identity.
-
Preclinical
- Research before or outside human clinical testing, often involving laboratory or animal models.
-
Purity
- A method-dependent assessment of the intended substance relative to detected material or impurities. Read the method and definition.
-
Randomization
- Assignment using a chance-based method to reduce systematic differences between groups.
-
Receptor
- A molecular target that recognizes a signal and can influence cellular activity.
-
Reconstitution
- Adding an appropriate liquid to a dried preparation.
-
Stack
- A grouping of compounds proposed for use together. The label does not establish that the combination has been studied.
Further reading: ClinicalTrials.gov glossary
Read terms in their study context
A word can be technically correct while still encouraging the wrong conclusion. For example, a response may refer to a hormone peak, a symptom score, or a tissue measurement. Whenever a paper reports a response, attach the measured endpoint and comparison. Likewise, activity may mean an effect in an assay rather than a useful whole-organism outcome. This glossary is a reading aid; the study's own definitions determine how its measurements should be interpreted.
Three distinctions to keep beside the glossary
Identity asks what material was studied. Evidence asks what happened when it was studied under defined conditions. Interpretation asks how far that finding can reasonably be applied. A chemical identity record cannot supply an effectiveness result, and an effectiveness paper cannot verify an unrelated supplier's batch. When you encounter an unfamiliar term, place it in one of these three questions before drawing a purchasing or research conclusion.
Read a study in context · Understand testing documents · Explore the compound library