How to Read a Certificate of Analysis (COA) for a Research Peptide

Open a box of research peptide and you are holding a small act of faith. Inside a glass vial is a few milligrams of white powder. It is, supposedly, a precise chain of amino acids in a precise order at a precise purity. But you cannot see any of that. Powder is powder. A truncated peptide missing one residue looks identical to the real thing; a vial that is 80% target and 20% process junk looks exactly like one that is 99.5%. The label tells you what someone hopes is inside. A Certificate of Analysis — a COA — tells you what a laboratory actually measured. It is the single document that turns “trust me” into “here are the numbers,” and learning to read one is the most useful, least glamorous skill in materials handling for Research Use Only work.

What is a Certificate of Analysis, really?

A COA is a structured laboratory report that documents the testing of one specific batch of material against a defined set of specifications.³ Think of it less as a certificate in the ceremonial sense and more as a lab notebook page made formal: here is the substance, here are the tests we ran, here is the method for each, here is the acceptance criterion, here is the result, and here — crucially — is the batch it describes. The international vocabulary for this comes from the ICH Q6A framework on specifications, which defines a specification as a list of tests, references to analytical procedures, and acceptance criteria that the material must meet to be deemed acceptable.³ A COA is the filled-in answer sheet to that exam.

The temptation is to skip to the big percentage at the top and move on. Resist it. A COA has a small number of sections, and each one answers a different and specific question about the powder in your hand. The two that matter most are identity and purity — and they are not the same thing, a distinction that trips up more people than almost anything else in this field.

How do you confirm identity — is it actually the right molecule?

Purity tells you how much of the sample is one dominant species. It does not tell you that the species is the one you ordered. You could, in principle, have a beautifully pure batch of the wrong peptide. That is what the identity section guards against, and the workhorse method is mass spectrometry. Every peptide has a characteristic molecular weight set by its exact sequence; the mass spectrometer weighs the molecule and checks whether the observed mass matches the theoretical one. A match within the expected tolerance is strong evidence that the backbone is the sequence the label claims.¹ Better COAs report the observed and theoretical masses side by side so you can check the arithmetic yourself, and analytical characterisation of synthetic peptides — confirming both identity and the impurity profile — is exactly what the EMA’s 2026 synthetic-peptide guideline asks manufacturers to demonstrate.¹

Mass alone has a blind spot worth knowing: two different sequences can occasionally share a near-identical mass, and some isomers weigh the same. That is why the strongest identity sections pair the mass result with orthogonal evidence — the HPLC retention time of a reference standard, or sequence-level data. For the compound primers we keep on this journal, from BPC-157 to Epitalon, identity is the first thing that should be nailed down, because everything downstream — every experiment, every comparison between labs — assumes the molecule is what it says it is.

What does the purity number actually mean?

Purity is the figure people quote, and it is measured by high-performance liquid chromatography (HPLC). The technique pushes the dissolved sample through a column that separates molecules by how they interact with the packing material; species come off at different times and register as peaks on a trace. The area under the main peak, expressed as a percentage of the total peak area, is the purity. A clean batch shows one tall, dominant peak and a flat baseline. The benchmark the field treats as the working quality threshold for research peptides is ≥99% by HPLC — a convention rather than a single regulatory number, which is why the more telling question is not the headline figure but what accounts for the rest.

Now the part most readers miss: the interesting information is in the small peaks. Every minor peak beside the main one is an impurity — and in synthetic-peptide chemistry these are not random dirt. They are predictable cousins of the target: deletion sequences missing a residue, truncated chains where synthesis stalled, peptides where a protecting group failed to come off, oxidised or deamidated variants, residual reagents.⁴ A purity figure of 99% means roughly 1% of the material is some combination of these. Whether that 1% matters depends entirely on what it is and what you are doing — which is precisely why impurities get their own reporting logic.

How are impurities supposed to be reported?

This is where a COA stops being a single number and becomes a document with a conscience. The ICH Q3A(R2) guideline, the international reference for impurities in drug substances, sets out a three-step logic that the peptide world borrows: impurities above a reporting threshold should be listed; above a higher identification threshold they should be chemically identified; and above a qualification threshold their safety should be assessed.² The EMA’s 2026 guideline applies this thinking specifically to synthetic peptides, where the impurity landscape is its own discipline and a reporting threshold in the region of 0.1% is the working convention.¹ A serious COA reflects this: it does not just say “99% pure,” it accounts for the rest. A weak one leaves the missing percentage in shadow.

What each COA section tells you — and how to tell a thorough certificate from a decorative one. Identity and the impurity reporting logic follow EMA and ICH characterisation guidance;¹² the test-and-specification structure follows ICH Q6A.³

What about water, appearance and the other tests?

Beyond identity and purity sit the quieter sections, and they earn their place. Water content — usually measured by Karl Fischer titration — matters because lyophilised peptides are hygroscopic; a vial that is 10% water by mass contains 10% less peptide than the label net weight implies, which quietly skews any concentration you calculate from it. Appearance sounds trivial until you remember it is a free, fast integrity check: a peptide described as a white powder that arrives discoloured or clumped is telling you something happened in handling or storage. ICH Q6A lists exactly these kinds of universal tests — description, identification, assay, impurities — as the backbone of a specification for any new substance.³ Some COAs go further with residual solvents or, for material destined for sensitive assays, endotoxin testing — relevant because peptides like BPC-157 carry real formulation and stability challenges that a thorough quality dossier is meant to surface.⁴

Why does the batch number matter so much?

Here is the idea that separates people who understand COAs from people who collect them. A COA describes one lot, not a brand. Synthesis is a physical process; two batches of the same peptide made weeks apart can differ in purity and impurity profile. So a COA is only meaningful when it is tied to the exact material in front of you by a batch (or lot) number and a date, and ideally a matching reference on the vial. A certificate for “Peptide X, our standard product” with no batch identifier is describing an abstraction. The ICH specifications framework is built on the premise that you test batches and release batches — the unit of quality is the lot.³ This is also the simplest test of whether a COA is real: does its batch number match the one on your vial, and does the date make sense?

How do you spot a weak or fake COA?

The tells are consistent. A weak COA is generic — it could apply to any vial of that peptide ever made, which means it applies to none of them. It omits the batch number and date, or carries ones that do not match your material. It states results without naming the methods (“purity: 99%” with no mention of HPLC, no chromatogram, no retention time). It reports identity as a bare “confirmed” with no observed mass to check. It shows a purity figure but is silent on the missing percentage, ignoring the impurity-reporting logic that ICH and EMA spell out.¹² And it lacks the connective tissue of a real document: who tested it, when, against what acceptance criteria. None of these red flags require a chemistry degree to catch — they require only the habit of asking, of each line, “does this describe my vial, and can I verify it?”

How much should you actually trust a COA?

Honestly: a COA is evidence, not a guarantee, and it is worth being clear-eyed about its limits. A certificate is only as trustworthy as the laboratory that produced it; a number can be transcribed, a chromatogram can be borrowed from a better batch, a method can be named without being run properly. The document does not test itself. This is why independent or third-party analysis exists, and why the most rigorous quality dossiers — like the formulation and quality discussion for BPC-157 in the recent Pharmaceutics literature — treat identity, purity and stability as an ongoing characterisation problem rather than a one-line claim.⁴ A COA does not make a vial trustworthy. It makes it checkable. The trust comes from a chain: a credible lab, methods you recognise, numbers that match your material, and the option to verify independently. A COA is the first and most important link in that chain — not the whole chain.

Why this matters for Research Use Only work

All of this is in service of one unglamorous goal: reproducibility. A research result is only as solid as the material it rests on. If you cannot say with confidence what was in the vial — the right molecule, at a known purity, from a documented batch — then a failed experiment and a contaminated reagent are indistinguishable, and a positive result cannot be told apart from an artefact. The Epitalon story we have written about elsewhere is a cautionary case of exactly this: confusing a pure peptide with a crude extract turned a translational error into decades of repeated mistakes. Identity by mass spectrometry, purity by HPLC, impurities held to a transparent threshold, water and appearance documented, all pinned to a specific lot — that is what a COA is for, and it is why every batch of Research Use Only material should arrive with one. These compounds are reference materials for laboratory research, not medicines; they are not for human or veterinary use, and a COA is a statement about quality and identity, never about use. Read this way, the certificate is not paperwork. It is the difference between a number you can build on and a powder you can only hope about.