How peptide COAs work — and how to read one

A certificate of analysis (COA) is the document that turns a peptide vial into a reproducible research reagent. It tells you what is in the vial, how the lab measured what is in the vial, and which lot the measurement was made against. This piece walks through each section of a typical research-peptide COA, what the numbers mean, and the common ways the document gets faked or oversold.

Section 1 — Identity

The top of the COA names the compound, its sequence (for peptides, this should be the full amino-acid sequence in single-letter code), and the molecular formula. For a 15-residue peptide like BPC-157, this is H-Gly-Glu-Pro-Pro-Pro-Gly-Lys-Pro-Ala-Asp-Asp-Ala-Gly-Leu-Val-OH, which compresses to GEPPPGKPADDAGLV. The expected molecular weight is calculated from the sequence and reported. For BPC-157 the monoisotopic mass is 1419.55 Da; if the COA shows a different number, either the sequence is wrong or the molecular-formula calculation was sloppy.

Section 2 — HPLC purity

The HPLC section reports purity as a percentage and shows the chromatogram. The chromatogram is the actual data; the percentage is a summary statistic computed from it.

A high-quality HPLC chromatogram for a synthetic peptide has a sharp, narrow main peak with most of the area concentrated under it, plus a clean baseline with small impurity peaks visible at expected retention times for deletion sequences and truncation products. A "≥99% purity" claim is meaningful when the chromatogram shows one dominant peak and the impurities are small, late-eluting, and well-resolved.

Red flags on the HPLC section:

• Purity claim with no chromatogram. The percentage without the visual is unverifiable.
• Broad main peak. Usually indicates incomplete deprotection or aggregation. The peak area integrates high but the compound is heterogeneous.
• Co-eluting impurities. A peak that has a shoulder is two compounds, not one. Integration captures both as "main peak."
• Different chromatograms for different lots. If every lot's chromatogram looks identical, the lab is probably using the same template image. Lot-specific data shows lot-specific variation.

Section 3 — Mass spectrometry (LC-MS or MALDI)

The mass-spec section confirms identity. It reports the observed mass and compares it to the theoretical mass from the sequence. For a clean synthesis, the observed and theoretical masses match within the instrument's tolerance (typically < 0.5 Da for time-of-flight instruments, < 5 ppm for high-resolution instruments).

Mass spec catches things HPLC cannot. Deamidation of asparagine residues shifts the mass by +1 Da. Oxidation of methionine shifts by +16 Da. Incomplete side-chain modification on engineered peptides like semaglutide shifts by hundreds of Da. None of these would show up clearly on an HPLC chromatogram of a 31-residue peptide, but they're trivial to see on mass spec.

For complex modified peptides — anything with a fatty-acid side chain, a covalent linker, or a non-standard amino acid — mass spec is the only realistic identity confirmation. HPLC alone is insufficient.

Section 4 — Water content (Karl Fischer)

Peptides are hygroscopic. A lyophilized vial labeled "5 mg" might contain 4.7 mg of peptide and 0.3 mg of residual water if not properly dried. Karl Fischer titration measures the water content directly. For research-grade material, water content under 5% is acceptable for most uses; under 2% is excellent.

Why this matters: if your stock concentration calculation assumes 5 mg of peptide and you actually have 4.5 mg, every dilution downstream is 10% off. For nM-scale receptor binding assays, that is enough to shift the dose-response curve visibly.

Section 5 — Counter-ion and salt form

Most synthetic peptides ship as a salt — trifluoroacetate (TFA) is the most common from standard solid-phase synthesis, acetate is preferred for some applications. The COA should specify which salt form is in the vial. If TFA content is reported (it should be, typically by ion chromatography), watch for it: TFA can interfere with some downstream assays.

Section 6 — Lot number and date

The lot number on the COA must match the lot printed on the vial label. A generic COA referenced by "see attached" or "available on request" without a lot link is not lot-traceable. The date of analysis should be within the production window — a COA dated months before the vial was synthesized is suspicious.

Section 7 — Lab identity

"Independent third-party lab" is marketing language. A real third-party COA names the analytical lab — its address, its CLIA or ISO 17025 accreditation status if applicable, and ideally a signed analyst. Anonymous "independent" labs are not third-party.

What good COAs look like

X Factor publishes per-lot COAs at /coa/. Each shows the HPLC chromatogram with the integrated peaks, the LC-MS spectrum with observed-vs-theoretical mass, the lot number matched to the vial, the contract lab named, and the date of analysis. The library is browsable before purchase, which is the point of publishing it. Most vendors keep COAs behind a "request" form because not all of them survive close reading.