How I Read a Peptide COA Chromatogram Without Pretending It Says More Than It Does
By Marcus Reid — Wed Sep 02 2026
How I Read a Peptide COA Chromatogram Without Pretending It Says More Than It Does — my honest, first-person take, backed by data from the 287 peptide vendors I track. Research use only.
I learned to read a peptide COA chromatogram the hard way: by comparing what vendors claimed, what the raw traces actually showed, and what later experiments confirmed (or contradicted). If you want to stop pretending a single peak proves everything, here’s how I walk through an HPLC peptide test and what I actually trust.
## Why I care — and what I don’t assume I track 287 vendor profiles because context matters: one chromatogram from one lot is meaningless without method details or a named lab. In my database only 65 of those 287 vendors (23%) publish named‑lab COAs — and that’s the first red flag I look for. A tidy, sharp main peak looks reassuring on a PDF, but it’s not evidence that the peptide is what the vendor says or that the sample lacks non‑UV‑active contaminants.
## The simple things I check first 1. Is the COA tied to the lot number and date? If not, stop. 2. Does the COA show method details: column, solvents, gradient, flow, wavelength? If not, stop. 3. Is it a named lab COA (third‑party)? If yes, that bumps my confidence. If no, I treat it as vendor‑reported data and probe harder.
These three checks eliminate half the false confidence I see in vendor listings. If you want a quick browse of products or vendors I compare, I often start at /peptides-list and /vendors; for practical calculations I use /peptide-calculator.
## My CLEAR checklist for reading a peptide COA chromatogram I use a short, repeatable framework I call CLEAR — it’s 5 concrete steps I run through every time.
- C — Credentials: Confirm lot number, analyst, and whether a named lab produced the COA. - L — LC method: Read column type, solvent system (TFA vs FA), gradient, flow, and UV wavelength. - E — Examine peaks: Look at main peak shape, retention time, shoulders, and baseline noise. - A — Ask for orthogonals: LC‑MS or MS/MS, amino‑acid analysis, or a peptide mass spec trace. - R — Record red flags: mismatched retention time, missing method, odd baseline, or unexplained big impurity peaks.
Follow CLEAR and you’ll stop being fooled by pretty PDFs.
## How I read the trace — step by step - Retention time (RT): I don’t care about an RT in isolation. I compare the RT to the vendor’s stated RT for that lot, and — ideally — to an authenticated standard run on the same method. RT shifts happen with columns, gradient slopes, or solvent modifiers (TFA vs formic acid), so a single RT number is weak evidence without method matching. - Peak shape: A single, symmetric Gaussian peak is what vendors want you to see. I watch for shoulders (co‑elution of isomers or impurities), broadness (heterogeneity or aggregation), and severe tailing (strong secondary interactions). These features tell me whether the peak truly represents a single species. - Integration and purity %: Percent area is method‑dependent. Integration ignores non‑UV‑absorbing material (salts, water, metals). A 95% area at 214 nm doesn’t mean “95% safe” — it means 95% of the UV‑absorbing material under those integration rules. Ask what the integration window and baseline algorithm were. - Baseline and noise: A drifting baseline or big solvent peak near the main peak can hide impurities. I zoom the baseline on the PDF and ask for the raw file if something looks suspect — a screenshot can be deceptive. - Wavelength: Most peptides are reported at 214 nm (backbone absorption). If the peptide contains aromatics, they may also show at 280 nm. If the COA only lists one wavelength, I ask why.
## What the HPLC peptide test does and doesn’t tell you - It reliably shows relative UV‑active purity under the method used. - It does not prove chemical identity. For that you need LC‑MS or MS/MS. - It does not reveal non‑UV contaminants (salts, heavy metals, endotoxin). - It can be gamed by poor integration, omission of method details, or by selecting a gradient where impurities co‑elute with the main peak.
If a vendor refuses to supply LC‑MS and insists a chromatogram is enough, I downgrade their trust level.
## A counter‑angle: “A crisp single peak = purity” is too simplistic The common advice I hear — “if it’s a single sharp peak, it’s pure” — is often repeated like gospel. I push back. A single UV peak can hide sequence isomers, deamidation products, or non‑UV impurities. For many routine research uses a well‑documented HPLC COA from a named lab may be adequate, but for new sequences, peptides that will be used in structural assays, or where downstream biology is sensitive, I demand LC‑MS. Rather than treating LC‑MS as mandatory for everything (which can be costly), I recommend a tiered approach: HPLC COA + named lab for low‑risk work; add LC‑MS/MS and amino‑acid analysis for critical or novel experiments.
## Practical red flags I always call out to vendors - No column type, gradient, or wavelength listed. - COA doesn’t include lot number and date. - Main peak has a visible shoulder or big nearby impurity over 5%. - Integration boundaries are cropped in the PDF (suggesting manual editing). - No orthogonal data (LC‑MS) available on request.
If a vendor can’t answer these within 48 hours, that vendor drops several places in my internal ranking.
## When the COA is honest — and what I do with it When I see a named‑lab COA tied to a lot, with method details, symmetric main peak, and a matching LC‑MS confirming the expected m/z, I move from suspicion to cautious acceptance. I still log the lot into my vendor profile and note whether the vendor routinely posts these documents; remember, only 65 of the 287 vendors I track publish named‑lab COAs, so consistent transparency matters.
If you want to compare how vendors handle documentation, check /vendors; for planning your experiments I keep a short spreadsheet of acceptance criteria based on CLEAR and method reproducibility.
*I’m not a doctor. This is how I, Marcus Reid, read peptide COA chromatograms for research‑use‑only purposes and help people separate robust evidence from vendor marketing. If you want a copy of my CLEAR checklist in a printable format, say the word and I’ll paste it out.*
*This article is for educational and research‑use‑only purposes. It is not medical advice and should not be used as guidance for human use.*
Frequently asked questions
What are the first things I look for when I open a peptide COA chromatogram?
When I open a chromatogram I first check that the main peak’s retention time matches the method’s reference and that the peak looks well shaped — sharp and symmetrical rather than ragged or heavily tailing. I note the baseline stability (no drifting or large noise spikes) and the presence, number and relative area of any additional peaks instead of pretending a single dominant peak proves everything. I remind myself that detector response and integration settings influence area%, so I treat the chromatogram as one piece of evidence, not a full verdict on identity or absolute purity — corroborating mass spec or orthogonal data is usually necessary. for educational and research-use-only purposes; this is not medical advice and no content should be treated as guidance for human use.
Can I rely on the chromatogram alone to confirm peptide identity and purity?
No — I don’t rely on it alone. A clean chromatogram is reassuring but not definitive: co‑eluting species can hide under the main peak and UV response varies with residue composition, so area% is a relative signal, not an absolute purity measure. When I need confidence I ask for orthogonal confirmation (mass spectrometry, intact mass or MS/MS, and, when relevant, an independent analytical method). I also check the method details and raw chromatogram if possible, because integration choices and processing can change the apparent result. Use the chromatogram as part of a weight‑of‑evidence approach, not a solo authority. for educational and research-use-only purposes; this is not medical advice and no content should be treated as guidance for human use.
What common mistakes do I guard against so I don’t overinterpret a COA chromatogram?
I avoid a few predictable traps: assuming a single peak equals one chemical species; treating reported area% as absolute purity without knowing detection wavelength and response factors; trusting heavily smoothed or baseline‑corrected traces without the raw data; and overlooking method limitations like poor resolution for hydrophobic or very polar impurities. I also don’t infer biological activity or safety from a chromatogram — it tells me about separation and signal, not function. When in doubt I request raw files, MS data, or a method blank so I can see what the chromatogram actually supports and what it doesn’t. for educational and research-use-only purposes; this is not medical advice and no content should be treated as guidance for human use.
References
About the author
Marcus Reid: Marcus Reid spent a decade in software engineering before going deep into peptide research, product documentation, and the clinical literature. He writes about what the data and the paperwork actually say. He is not a doctor; PeptideTally content is educational and does not constitute medical advice.