Peptide test reports often include a chromatogram beside a purity percentage. The percentage is easy to read, but the graph takes more interpretation.
The graph shows the detector signals behind the reported percentage, including which peaks were counted.
This guide explains the axes, peaks, area calculations and limits of an HPLC purity result.
What the instrument does to the sample
HPLC stands for high-performance liquid chromatography. A pump moves a dissolved sample through a column packed with particles that help separate its components.
Peptide purity testing commonly uses reversed-phase HPLC. In this method, components interact with the column differently. Those that interact less strongly pass through sooner; others come out later.
Other HPLC methods separate substances by properties such as size or electrical charge. The chromatogram still records detector response over time, but the separation works differently.
A detector measures substances as they leave the column. The chromatogram then plots that response over time.
The two axes
The horizontal axis shows time since the sample was injected, usually in minutes.
The vertical axis shows detector response. Peptide testing commonly uses ultraviolet (UV) detection, which measures how much light the liquid absorbs. The signal is often shown in milli-absorbance units (mAU).
A rising line means the detector is registering material at the selected wavelength. A flat trace, by contrast, means nothing is producing a detectable increase above the baseline. Material that does not absorb at that wavelength, or is present below the detection limit, may still be passing through.
What retention time tells you
Retention time is the clock reading at the top of a peak. The chapter on chromatography in the United States Pharmacopeia defines it as the “time elapsed between the injection of the sample and the appearance of the maximum peak response.”
Under consistent testing conditions, the same compound should emerge at roughly the same retention time. That repeatability makes the measurement useful for comparison.
Different compounds can share a retention time, and that time can shift when testing conditions change. So a matching retention time alone cannot establish the compound’s identity.
Peaks, height, and area
A peak appears when one or more components leave the column and are detected. A tall, narrow peak and a shorter, wider peak can have the same integrated area.
Laboratories commonly use peak area rather than height when calculating HPLC area percentages. Area measures the detector response between the trace and its baseline, taking account of signal intensity and duration. The United States Pharmacopeia, though, allows either area or height as a measure of peak response.
Peak height can still be useful, especially when a small impurity peak overlaps the tail of a much larger peak.

How peak areas become a purity percentage
A common calculation is area normalization. The laboratory totals the integrated areas of reportable peaks and divides the main peak’s area by that total. Depending on the method, solvent or matrix peaks and signals below the reporting threshold are excluded.
This illustrative peak table uses invented numbers for the calculation, not results from a laboratory test.
| Peak | Retention time | Area | Percent |
|---|---|---|---|
| Impurity A | 10.1 min | 61,300 | 0.613% |
| Impurity B | 11.8 min | 38,900 | 0.389% |
| Main peak | 12.4 min | 9,812,400 | 98.124% |
| Impurity C | 12.9 min | 87,400 | 0.874% |
| Total | 10,000,000 | 100.000% |
The three impurity areas total 187,600, and all four areas total 10,000,000. The main peak accounts for 9,812,400 divided by 10,000,000, or 98.124% of the counted detector signal. Rounded, that is 98.1%.
That is a percentage of the detected peak areas, not a measurement of how much peptide powder is in the vial.
Where the baseline goes changes the number
Before calculating peak areas, software identifies where each peak starts and ends and where to place its baseline. This process is called integration.
Integration is generally straightforward when peaks are well separated. When a small impurity overlaps the tail of a larger peak, the baseline choice can change the area assigned to each. Those choices are not necessarily equally accurate; the appropriate approach depends on peak shape and separation.
Suppose a different baseline assigns 40,000 units of impurity C’s area to the main peak instead.
| Peak | Area | Percent |
|---|---|---|
| Impurity A | 61,300 | 0.613% |
| Impurity B | 38,900 | 0.389% |
| Main peak | 9,852,400 | 98.524% |
| Impurity C | 47,400 | 0.474% |
| Total | 10,000,000 | 100.000% |
The sample has not changed, but the reported main-peak area rises from 98.1% to 98.5%.
In fact, published research has found larger errors. A study in LCGC measured those integration errors directly and found that when a small peak sits beside a much larger one, the reported area for the small peak can be off by more than 50%. The author concluded that most of the error comes from “extended tailing from the large peak,” and that resolution below 1.5 should be avoided in those cases. Measuring the small peak by height rather than area was consistently less error-prone.
A small impurity peak can have a large relative integration error even when the effect on the overall purity percentage is modest.
What the detector can and cannot see
A UV detector only registers substances that absorb ultraviolet light at the wavelength it is set to. Peptide work is commonly done at 214 nm, a wavelength where the bonds joining amino acids absorb strongly.
Material that does not absorb at the selected wavelength may go undetected and therefore be absent from the area-percentage calculation.
Even among things that do absorb, the response is not uniform. How strongly a peptide absorbs varies with composition, and work published in the Journal of Chromatography A found that peptides containing tryptophan behaved differently from those containing only tyrosine. As a result, two peaks of identical area can represent different amounts of material.
Area normalization assumes that the measured peak responses can be compared. Different compounds may respond differently to the detector, so a method may require correction factors. A report should explain whether such corrections were used.
Small peaks the report may leave out
Some signals are too small for a method to report as peaks. Laboratories specify a disregard or reporting threshold that determines which signals are included in the calculation.
There is no single number that applies everywhere. The general pharmaceutical guideline on impurities, ICH Q3A, is often quoted for a 0.05% reporting threshold, but its own scope excludes peptides by name, along with biological, oligonucleotide and fermentation-derived substances. So a threshold taken from it does not transfer to a peptide method.
Peptide-specific expectations, meanwhile, are being rewritten. The Food and Drug Administration withdrew its 2021 guidance on synthetic peptide generics in July 2026, saying it no longer reflected the agency’s current scientific thinking, and issued revised draft product-specific guidances the same day.
For the report in front of you, the important details are the method’s reporting threshold and how it was determined. A threshold borrowed from an unrelated guideline does not establish what this test could detect.
For applicable pharmacopeial procedures, system-sensitivity requirements can link the limit of quantitation to a stated reporting threshold. The limit is commonly associated with a signal-to-noise ratio of 10. Those criteria, though, depend on the procedure and should not be assumed to apply to every research-peptide method.
Without a stated threshold, readers cannot tell whether small impurity peaks were absent or simply not counted.
Checking system suitability
Laboratories use system-suitability checks to confirm that an analytical method is performing within its specified limits during testing.
Common measures include peak symmetry (tailing factor), separation between neighboring peaks (resolution) and consistency across repeated injections (repeatability).
Acceptance limits depend on the method. Pharmacopeial procedures may set specific criteria, while laboratories using validated in-house methods establish criteria suitable for those tests. The results should be evaluated against the criteria actually used.
A report containing only a chromatogram and a purity percentage does not provide enough information to assess system suitability. Still, that omission does not prove the test failed.
What a chromatogram cannot establish
A chromatogram used for purity testing answers a limited analytical question. It does not establish several other properties of the product.

It does not confirm identity. A peak at the expected time is consistent with the right peptide, not proof of it. The international guideline on analytical validation says that where a single procedure is not sufficiently selective, an additional procedure should be used, and in peptide work that usually means mass spectrometry.
It also does not prove a peak contains one substance. Two compounds can come off the column together and appear as a single peak. The Agilent and Novo Nordisk authors cited above needed mass detection to pull apart species that co-eluted, and noted that isomers cannot be separated by mass alone either.
It does not measure sterility, and it does not measure bacterial endotoxins. Those are separate tests on separate equipment, and a sterile product and a pure one are different claims.
Finally, it does not establish the peptide quantity in the vial or whether the product is safe for human use.
Reading the report around the chromatogram
A chromatogram is easier to evaluate when the report identifies the sample, method and calculations.
Check the laboratory’s name, the sample identifier, batch or lot number, and test date. Without these details, it may be impossible to connect the trace to the material being evaluated.
Check the method, including column, solvents, gradient, detection wavelength and injection volume. Different validated methods can yield different apparent purity percentages from the same material.
Then review the peak table, integration marks, reporting threshold and system-suitability results. Together they show how the laboratory derived the percentage and whether it documented method performance.
The report still cannot establish whether the vial in hand came from the tested lot, or how storage and shipping affected it afterward.
The chromatogram and its supporting details show what the laboratory measured, and what the percentage leaves unanswered.