Identity: why it comes first

Lesson 1 split every result into question, method, output and limit. This lesson takes the first question, "is it the labelled molecule?", and shows why it is answered before the others and how to judge the strength of an identity result.

Why must identity be settled before purity and amount?

Because purity and amount are both calculated relative to something that identity has to confirm.

• Purity is the main peak's area divided by the total detected area. The calculation picks the biggest peak and calls it the product. It never asks what that peak is. If the vial holds a different peptide, purity becomes a precise measurement of the wrong thing.
• Amount is the sample's peak compared with a reference standard of the labelled compound. If the sample is a different molecule, it will usually sit at a different place in the chromatogram and respond differently to the detector, so the milligram figure has no defined meaning. If an analyst simply integrates the biggest peak anyway, the number looks normal and is wrong.

This is why a report that leads with "99% pure, 10.1 mg" and buries or omits identity is backwards. The first two figures are conditional on the third.

What does "orthogonal" mean for an identity test?

Orthogonal methods measure different physical properties, so they fail in different ways. One method can be fooled by a molecule that happens to share the property it measures; two unrelated methods are much harder to fool at the same time.

The European Medicines Agency's guideline on synthetic peptides says that for identification "use of at least two orthogonal methods is recommended", and lists identification by mass, relative retention time, LC-MS, peptide mapping, bioactivity, amino acid analysis or NMR as appropriate (EMA, synthetic peptides guideline, 2025). It adds that the chosen test or combination must be "suitable to unambiguously confirm the sequence of the peptide".

Each method page covers the details: LC-MS, MALDI-TOF, peptide mapping and amino acid analysis.

When is a mass match not enough?

When two molecules have the same mass. The textbook case is leucine and isoleucine: they are isomers, with the same atoms arranged differently, so no mass measurement can separate them. A peer-reviewed paper on peptide reference standards describes using amino acid analysis "as an identity method to distinguish between leucine and isoleucine" for exactly this reason (McCarthy et al., Pharmaceutical Research, 2023).

ICH Q2(R2) builds this risk into validation: an identification test should be applied to "materials structurally similar to or closely related to the analyte to confirm that a positive result is not obtained" (ICH Q2(R2)). An identity test has only been shown to work if it says no to the near misses.

For the swaps buyers worry about most, mass alone is usually decisive. Using PubChem's monoisotopic masses, tirzepatide is 4810.52 Da (PubChem), retatrutide is 4728.47 Da (PubChem, CID 171390338) and semaglutide is 4111.12 Da (PubChem). The closest pair, tirzepatide and retatrutide, are about 82 Da apart. Lesson 7 shows how that gap appears on a spectrum.

These are not hypothetical swaps. The US FDA warns that illegally marketed semaglutide and tirzepatide "could contain the wrong ingredients", or "too little, too much or no active ingredient at all" (US FDA, concerns with unapproved GLP-1 drugs). The first of those is an identity question, and the same white powder in the same kind of vial gives no visual clue. The others are amount questions (lesson 4), which only have an answer once identity is settled.

What about a molecule that is almost right?

Some impurities are versions of the target with one amino acid missing or one extra. The EMA guideline calls these deletion sequences, formed by incomplete coupling or deprotection steps during synthesis, and insertion sequences, formed when an amino acid is coupled more than once in one step. They are near copies made in the same process, not swaps.

As a minor component, a near copy is a small chromatogram peak and a purity question (lesson 3). As the main component, it makes the vial's identity wrong even though the molecule is closely related. Intact mass catches most of these, because losing or adding a residue changes the mass by that residue's weight; sequence methods confirm where the change is.

How strong is the identity result in front of you? A worked example

These four identity lines are invented for teaching and describe no real report. Rank them from weakest to strongest.

The third line is what most buyers see on a good report, and for the common substitution risks it is enough. The fourth matches the two-method approach the EMA guideline recommends for release testing. When you have a report with expected and observed masses, the mass checker compares them for you against the published mass of the labelled compound.

What should you do if identity is missing or fails?

If identity is missing, treat every other number on the report as unconfirmed. Purity and milligrams still describe the sample's main component, but you do not know what that component is. If identity fails, the other numbers describe a different product from the one on the label and should not be used to compare value or quality against other vials of the labelled compound.

A failed identity on one batch says nothing about another batch from the same seller, in either direction. Next, with identity settled: Purity: what HPLC measures.

Check your understanding

1. A vial labelled retatrutide shows 99% HPLC purity and no identity result. What do you know about the molecule? Nothing. Purity shows the sample is mostly one compound; it does not say which.
2. Why can a mass match alone not separate leucine from isoleucine in a sequence? They are isomers with the same mass. A different method, such as amino acid analysis, is needed.
3. What makes two identity methods orthogonal? They measure unrelated properties, such as mass and chromatographic retention, so a molecule that fools one is unlikely to fool the other.