Testing, quality and handling
Peptide COA grading: a transparent checklist instead of a single score
NB
· 9 min read
A summary of published research, with citations. It states no amount, schedule or route, and makes no claim about what this compound does in a person. Supplied for laboratory research use only.
On this page · 7 sections
The most reliable way to grade a peptide certificate of analysis is a checklist that records each criterion separately rather than one letter or number. Start with three pass-or-fail gates: the report is confirmed by the issuing laboratory, its lot matches the vial, and its sample matches the product. Then score identity, purity method, peptide content and safety testing on their own lines, and record "not tested" as its own answer. A single combined grade lets a strong result in one area hide a missing one.

This page is the scoring tool. The reasoning behind each line lives elsewhere: confirming a report is genuine in how to verify a peptide COA, what each measurement means in what a certificate of analysis establishes, and judging the testing behind the report in third-party peptide testing in Canada.
Why single grades hide weaknesses
A summary score adds things up. Adding works when the parts are interchangeable, and the parts of a certificate are not. A detailed chromatogram cannot compensate for the absence of an identity test, because they answer different questions. A sample of the wrong molecule can produce an immaculate chromatogram.
Clinical research learned this the expensive way. When Jüni and colleagues applied 25 different published quality scales to the same 17 trials, the scales disagreed about which trials were high quality, and the conclusion of the meta-analysis changed depending on which scale was used. Summary quality scores were not significantly associated with treatment effects, while one individual methodological feature was. The authors' recommendation was to assess relevant aspects individually rather than rely on summary scores (PMID 10493204). The same reasoning applies to grading a certificate.
Single grades fail in four specific ways:
- Compensation. Points earned on purity offset points lost on identity or lot matching, so a report with a fatal gap can outscore a sound but modest one.
- Hidden weighting. Every total embeds a decision about how much each criterion is worth. If the weights are not published, nobody can tell whether a grade reflects the report or the grader's priorities.
- "Not tested" disappears. A missing endotoxin test scored as zero looks like a weak result; ignored entirely, it looks like nothing is missing. Neither is accurate.
- False precision. A grade of 87 out of 100 implies a resolution the underlying document does not have, and invites comparison between reports whose differences are not on the same axis.
Principles of a rubric that two people can reproduce
A useful rubric gives the same answer whoever applies it. That requires criteria that can be read off the document or checked against an outside record, rather than criteria that call for an impression.
- Gates before scores. Some failures make every other line meaningless. If the report cannot be confirmed with the laboratory, or describes a different lot, there is nothing left to grade.
- Observable anchors. Each level is defined by what is present on the page — "chromatogram with peak table" — not by a judgement such as "good method".
- Separate dimensions. Identity, purity, content and contamination are recorded on separate lines and never summed.
- Three-valued answers. For each test: reported, not reported, or failed. Absence is information.
- Written evidence. Each line records where the answer came from: the portal, the vial label, the page of the report.
The checklist
Each scored line has explicit anchors. Score what the document shows, not what the seller says it would show.
| Section | Criterion | How to check | Levels |
|---|---|---|---|
| Gate 1 | Report authenticity | Retrieve from the issuing laboratory's portal or confirm by email; compare every field | Pass / Fail / Unable to confirm |
| Gate 2 | Lot match | Lot on the certificate against the lot printed on the vial | Pass / Fail / No lot on vial |
| Gate 3 | Sample matches product | Compound name, stated sample mass, every component of a blend | Pass / Fail |
| Identity | Is the main component the intended molecule? | Look for mass spectrometry data | 0 none · 1 mass reported separately, observed matches theoretical · 2 LC-MS, mass assigned to the main peak |
| Purity method | Can the purity figure be audited? | Look for the trace and method details | 0 figure only · 1 chromatogram with peak table · 2 plus column, gradient, wavelength and system suitability |
| Content | How much peptide is in the vial? | Look for an assay or net peptide content result | 0 not reported · 1 reported without method · 2 reported with method |
| Endotoxin | Is lipopolysaccharide measured? | Look for a result in EU/mg | Reported / Not reported |
| Other tests | Water, counter-ion, residual solvents, sterility, elemental impurities | Look for each by name | Record each as reported or not reported |
| Accreditation | Was the method within the laboratory's accredited scope? | Accreditation body's directory and scope document | Yes / No / Not checked |
| Sample origin | Who supplied the sample to the laboratory? | Any statement on the report or programme page | Independent / Seller / Not stated |
| Timing | Does the test date fit the lot? | Test date against when the lot was offered | Consistent / Inconsistent |
A report that fails any gate is recorded as a failure and graded no further. A report marked "unable to confirm" at Gate 1 is not a failure, but nothing below it should be treated as more than the seller's claim.
How to record the result
Write the outcome as a profile, not a total. One line is enough:
Gates: pass / pass / pass · Identity 2 · Purity method 1 · Content 0 · Endotoxin not reported · Accreditation not checked · Sample origin not stated
That line can be compared across vendors without pretending the dimensions trade off. If you must rank, rank on the weakest dimension that matters to your work, not on a sum. For cell or animal work, an unreported endotoxin result may matter more than any refinement in the purity method. For analytical work, content may matter most. The checklist records the facts; the weighting belongs to the person whose experiment depends on it, and should be written down beside the profile.
A worked comparison
Consider two hypothetical reports for the same compound.
Report A shows a very high purity figure, a clean chromatogram with full method details and a laboratory logo. There is no mass spectrometry, no content result and no endotoxin test. The lot printed on it is not the lot on the vial.
Report B shows a lower but respectable purity figure, a chromatogram with a peak table and no gradient details, LC-MS with the observed mass assigned to the main peak, a content result with its method, and an endotoxin result. It has been confirmed with the laboratory, and its lot matches the vial.
A points-based grade weighted towards purity and presentation could rank Report A higher. The checklist stops at Gate 2 for Report A: whatever it measured, it did not measure this vial. Report B passes all three gates and scores Identity 2, Purity method 1, Content 2, Endotoxin reported. One of these documents tells you something about the material you hold. The other does not, and only a separated checklist makes that visible at a glance.
The scenario is not contrived. Researchers who ran their own quality control on synthetic peptides ordered at a specified purity found a large discrepancy with the suppliers' certificates; fewer than half met the specification, and one contained a different structure from the peptide ordered (Verbeke and colleagues, PMID 29403929). A grade built on the supplier's own figure would have scored every one of those certificates well.
What the checklist cannot score
Some questions need more than the document. A purity method can be fully described and still use a gradient too steep to separate closely related impurities, such as one-residue deletions or racemised residues, which are among the commonest products of solid-phase synthesis (D'Hondt and colleagues, PMID 25044089). Judging that needs someone who reads chromatograms; the HPLC and LC-MS explainer is a starting point.
Nor can any score describe the vial after its test date. Shipping temperature, storage and handling are outside every certificate. And no grade, however high, says anything about what a compound does in a biological system.
The certificates this site holds are listed by compound and laboratory on its lab results page. The checklist applies to them as to any others; for the wider buying decision, see buying research peptides in Canada.
Frequently asked questions
What is peptide COA grading? It is the practice of assessing a certificate of analysis against a set of criteria to judge how much it tells you. Done well, it records each criterion separately: whether the report is genuine, whether it matches your lot, and which tests it includes. Done badly, it collapses everything into one letter or number that conceals what is missing.
Why not just add the checklist scores up? Because the criteria are not interchangeable. A strong purity method cannot make up for an absent identity test, and a perfect report on the wrong lot describes nothing you own. Adding scores lets one strength offset an unrelated weakness, and the result depends on weights that are rarely disclosed or justified.
Should a missing test count as a fail? Record it as "not reported", a third value distinct from both pass and fail. A missing endotoxin test does not show contamination, but it does not show its absence either. Treating it as zero exaggerates the problem; ignoring it hides the problem. The profile should show the gap plainly.
Can I trust a vendor's own COA grade or quality score? Only as far as you can reproduce it. Ask what criteria the grade uses, how each is weighted, and whether the underlying reports have been confirmed with the issuing laboratory. A grade without published criteria cannot be checked, and a seller grading its own certificates is reporting on itself.
Which checklist line matters most? The gates. If a report cannot be confirmed with the laboratory, or names a different lot, nothing else on it applies to your material. After the gates, identity by mass spectrometry is usually the most informative single line, because it is the only test that shows the main component is the molecule on the label.
Selected Research References
- Jüni P, Witschi A, Bloch R, Egger M. The hazards of scoring the quality of clinical trials for meta-analysis. JAMA. 1999;282(11):1054-1060. pubmed.ncbi.nlm.nih.gov/10493204
- Verbeke F, Wynendaele E, Braet S, et al. Quality evaluation of synthetic quorum sensing peptides used in R&D. J Pharm Anal. 2015;5(3):169-181. pubmed.ncbi.nlm.nih.gov/29403929
- D'Hondt M, Bracke N, Taevernier L, et al. Related impurities in peptide medicines. J Pharm Biomed Anal. 2014;101:2-30. pubmed.ncbi.nlm.nih.gov/25044089
All products sold by NovaEvum are for research and laboratory use only. They are not intended for human or animal consumption, diagnosis, treatment, cure, or prevention of any disease.
