How should Australian labs read HPLC results on a peptide COA peak table?
An HPLC peak table is a structured extract of the chromatogram. On research-peptide certificates of analysis circulated in Australia it is commonly presented as a grid beneath, or on the page following, the plotted trace. Typical columns include peak number, peak name or identifier, retention time, relative retention time against a designated principal peak, area in instrument-native units, area-percent, height, and sometimes width at half-height or a tailing factor. Not every chromatography data system exports every column. The reader's first task is to catalogue which columns are populated and which are blank, because a missing relative-retention-time column changes how related substances can be compared with a specification.
Identity of the record comes before interpretation of the numbers. The chromatogram header should carry the same lot number as the COA, the same sample identifier as the labelled vial or aliquot, and a method name or method version that can be traced to a controlled analytical procedure. Sequence-table position, data-file name and acquisition date-time form part of that identity block. Mismatches among header, peak table and COA cover sheet are documentation defects; they are not resolved by averaging purity figures across pages.
The principal peak is the signal assigned to the target sequence. Its retention time should fall inside the identity window defined by the method. Related substances, if named, should be labelled consistently with the method's peak-naming table. Deletion sequences, insertion sequences, methionine sulphoxide species, aspartimide species and incompletely deprotected residues are typical synthetic-peptide related substances. Unnamed peaks should still carry a retention time and an area so that they remain available for summation. Peptide bonds are commonly monitored near 214-220 nm; aromatic side chains may be monitored near 280 nm. A peak table generated at 280 nm will under-represent sequences that lack aromatic residues, so the wavelength printed on the chromatogram must match the method that underpins the specification. Inspect integration events on the plot, because start and stop marks, tangential skim versus drop-line, and valley-to-valley splits determine the areas that later become area-percent.
How is HPLC area-percent purity reconstructed from the peak table?
Area-percent purity is not an independent measurement; it is a ratio calculated from the same areas printed in the peak table. In the simplest convention, the area of the principal peak is divided by the sum of all integrated peak areas assigned to the sample, then multiplied by one hundred. If the table already contains an area-percent column, the principal-peak cell should equal that ratio within ordinary rounding. Disagreement between a cover-sheet purity-by-HPLC value and the peak-table area-percent is a reconciliation failure and should be queried with the issuing laboratory before the lot is accepted into a research inventory.
Two arithmetic details cause most discrepancies. First, rounding: individual area-percent values may be reported to two decimal places while the cover sheet is rounded to one decimal place. Reconstructing the ratio from raw areas, rather than summing rounded percentages, is the more defensible check. Second, the denominator: some methods sum only peaks above a reporting threshold; others sum every integrated event including those later flagged as solvent. If the COA does not state the summation rule, the peak table cannot be said to support the headline number. Height-percent is a different statistic and is not interchangeable with an area-percent specification.
A methodical reading of tabulated experimental results—checking definitions of each statistic before accepting a summary figure—is a long-standing recommendation in the information-science and medical-reading literature. Richmond's notes on how to read experimental material emphasise attending to the structure of tables rather than to captions alone (DOI:10.1002/asi.5090180408). Darzins and colleagues, writing in the Medical Journal of Australia, similarly advise a structured pass through methods and data before conclusions (DOI:10.5694/j.1326-5377.1992.tb137249.x). Applied here, that means: write down the summation rule, recompute the ratio from areas, and only then compare the result with the specification limit printed on the COA. Australian procurement files should keep the peak table with any orthogonal identity page so that later readers do not treat a single headline as a complete characterisation.
Which HPLC peaks should be excluded as solvent, diluent or system peaks?
Solvent-front and diluent peaks are signals that arise from the sample solvent, the mobile-phase mismatch at the void, or a system peak that appears in blank introductions of the same diluent. They are not related substances of the peptide. If they remain inside the summation used for area-percent, the principal-peak percentage is diluted and the purity figure is no longer a statement about the peptide-related peak set.
A defensible COA therefore states an exclusion rule. Common rules include: exclude all peaks with retention time below a stated multiple of the void time; exclude peaks that are present at comparable area in a blank chromatogram acquired in the same sequence; exclude named diluent peaks listed in the method. The peak table should flag excluded peaks so that a later reader can reproduce the denominator. If excluded peaks are simply omitted from the table, the reader cannot test whether a large unassigned signal was dropped without justification.
Blank chromatograms belong with the batch file. A blank that is noisy at the principal-peak retention time undermines both identity and purity, because a system peak could be contributing area to the principal signal. Baseline slope and drift should be visible on the plotted chromatogram; a severely sloping baseline with drop-line integration can transfer area from a tailing principal peak into adjacent impurity peaks or the reverse.
Sample-preparation fields on the COA—diluent identity, nominal concentration, and chromatographic sample-introduction volume—explain why a solvent peak is large or small. A peptide-related early eluter can sit near the void; excluding everything before a crude time cut may hide a real deletion sequence. The method's exclusion window should therefore be justified in the procedure, not improvised on the chromatogram. Australian laboratories comparing suppliers should ask whether blank, system-suitability and sample chromatograms are issued as a set. Local AU stock with batch documentation and tracked dispatch makes it practical to keep that set together under one lot identifier.
How do relative retention time and peak names support lot traceability?
Relative retention time (RRT) is the retention time of a given peak divided by the retention time of the principal peak in the same chromatogram. Because absolute retention time moves with column age, temperature and mobile-phase preparation, RRT is the more portable identifier for related substances across sequences and, with caution, across equivalent columns. A peptide COA that lists related substances only as impurity 1 and impurity 2 without RRT forces the receiving laboratory to re-establish identity of each minor peak from scratch.
Lot traceability requires a chain of identifiers, not a single purity number. The vial label, the COA header, the chromatogram header, the data-file name and the packing list should share one lot code. If a peak-naming table is used, the names on the chromatogram should match the names in the specification, for example Met(O) for a methionine sulphoxide or des-Gly for a terminal deletion. Inconsistent naming across pages of the same PDF is a documentation defect.
Finding the right field in a multi-page analytical packet is itself a documentation problem. Cavière's account of documentation for Atlas of Living Australia tools stresses that users must know how to locate the correct information object among many similarly labelled resources (DOI:10.3897/tdwgproceedings.1.19941). A research-peptide batch file is analogous: the HPLC peak table, the mass-spectrometry spectrum, the residual-solvent page and the net-content page are different objects. Lot traceability fails when those objects are filed under different informal filenames.
RRT windows should be stated. A related substance specified at RRT 0.86 with a window of 0.02 is a testable claim; a name without a window is not. If two related substances have RRTs closer than the method's resolution allows, the peak table may show a single fused peak, and the area cannot be attributed to one named impurity. For multi-vial orders drawn from one lot, each vial label must still point to the same lot's peak table. Australian buyers comparing wholesale packs should verify that every vial's lot code resolves to one HPLC file; that check is a supplier-evaluation step in a research-only supply chain.
How do reporting thresholds, method identifiers and sign-off blocks change HPLC COA reading?
Chromatography data systems do not list every deflection on the baseline. A peak is entered into the table only when it meets integration criteria: a minimum area, a minimum height, a minimum width, or a signal-to-noise ratio. Those criteria are method parameters. If they are omitted from the COA, two laboratories can process the same raw file into different peak tables. The reporting threshold is the level at which a peak must be listed and, separately, the level at which it must be quantified. Without a stated rule, an unidentified peak that is absent from the table may be either truly negligible or simply below an undisclosed cut-off. Aggressive smoothing can merge a shoulder into the principal peak and inflate area-percent; those choices should be visible as integration events and summarised in the method printout.
Interpreting a results table requires knowing what each reported statistic was designed to represent. Commentary on how tabulated survey results should be read for a new context makes the same point in a different domain: a figure is not self-explanatory without its definition and its intended use (DOI:10.7816/ulakbilge-05-15-06). On a peptide COA, an unidentified peak at 0.12 per cent means an integrated signal that met the reporting rule, was not matched to a named related substance, and contributed 0.12 per cent of the chosen denominator. It does not mean a confirmed molecular structure. If the specification limits unspecified related substances, the reader must sum only unspecified peaks and must use the same denominator as the specification.
A peak table is only as trustworthy as the method that produced it. The COA should carry a method identifier and version, column identity, mobile-phase composition, gradient programme or isocratic proportions, flow rate, column temperature, detection wavelength and bandwidth, and the sample-introduction volume. System-suitability pass/fail against stated limits should be explicit. Sign-off blocks—an analyst identification, a reviewer identification, and date-time stamps—indicate that the integration was inspected. How-not-to-read commentary in another literature warns against imposing a preferred narrative on an incomplete document (DOI:10.5840/acpq199569241). The laboratory analogue is treating an unsigned, headerless chromatogram as if it were a reviewed batch record. Australian laboratories should not interpret through a method-version mismatch. Before a research-only purchase is closed, check that method identifiers, lot codes and sign-off blocks are complete, and that local AU stock can be sent with that packet under tracked dispatch. That is a documentation and supplier-evaluation test; it does not characterise biological activity.
Apply this checklist to documented stock
You now have a practical way to read purity figures, method notes, and lot traceability. When you source materials, hold suppliers to that same checklist — ClaraScience issues batch documentation with every order and dispatches from Australian warehouses with Express tracked shipping.
Start with a retail order to review documentation end-to-end, or register for wholesale if you restock multiple compounds.
Frequently asked questions
What should be checked first on peptide COA HPLC documentation in Australia?
Match lot number, sample identifier and method identifier across the vial label, COA header and chromatogram header. Then confirm that the peak table belongs to that file (data-file name, sequence position, date-time). Only after identity concordance should area-percent be reconstructed. Research use only; this is a documentation check, not a biological characterisation.
Why might HPLC area-percent on the peak table disagree with the COA cover sheet?
Rounding, a different denominator, exclusion of solvent peaks, or substitution of height-percent for area-percent are the usual causes. Recompute the ratio from raw areas using the method's stated summation rule. If the rule is not stated, the headline purity is not auditable. Query the issuing laboratory before accepting the lot into a research inventory.
Should solvent-front peaks be included in peptide HPLC purity?
No. Solvent, diluent and blank-matched system peaks are not peptide-related substances and should be flagged as excluded in the peak table. The exclusion window must be defined in the method so that an early-eluting deletion sequence is not discarded with the void. A blank chromatogram from the same sequence is required to verify the decision.
How does relative retention time support peptide lot traceability?
RRT normalises minor-peak retention to the principal peak in the same chromatogram, which is more stable than absolute retention time when column age or mobile-phase preparation shifts. Named related substances should carry an RRT and a window. Lot traceability still requires the same lot code on the vial, COA, chromatogram and packing list.
What HPLC method fields should an Australian laboratory expect on a peptide COA?
Method identifier and version, column identity, mobile-phase composition, gradient programme or isocratic proportions, flow rate, column temperature, detection wavelength and bandwidth, sample-introduction volume, system-suitability pass/fail, and analyst/reviewer sign-off. Missing fields mean the peak table cannot be tied to the specified procedure.
Are these HPLC COA checks a substitute for in-house identity testing?
They are documentation and supplier-evaluation checks for research-use-only materials. A laboratory quality system may still require orthogonal identity confirmation on receipt. Reading the COA does not characterise biological activity and does not support human use.
References
- DOI:10.1002/asi.5090180408 — Suggestions on how to read experimental material in information science — American Documentation — 1967
- DOI:10.5694/j.1326-5377.1992.tb137249.x — 13. How to read a journal article — Medical Journal of Australia — 1992
- DOI:10.3897/tdwgproceedings.1.19941 — Documentation about Atlas of Living Australia tools: how to find information — Proceedings of TDWG — 2017
- DOI:10.7816/ulakbilge-05-15-06 — HOW TO BE READ PIAAC RESULTS FOR ADAPTATION TO NEW AGE? — Ulakbilge Dergisi — 2017
- DOI:10.5840/acpq199569241 — How (Not) To Read Heidegger — American Catholic Philosophical Quarterly — 1995
Research use only
This article is provided for laboratory research and educational purposes only. Products referenced are not for human or veterinary use. ClaraScience makes no therapeutic, medical, or efficacy claims, and nothing here constitutes medical advice.
