How to read a peptide Certificate of Analysis

The top of a peptide COA identifies the document and the material it covers. Look for:

• Issuing laboratory — the name and address of the organization that performed the testing. For an independent verification, this should not be the same entity that manufactured the peptide.
• Product name and sequence — the peptide identifier (common name, INN, or systematic name) and the full one-letter or three-letter amino acid sequence.
• CAS number and molecular formula — chemical identifiers for unambiguous identification.
• Theoretical molecular weight — calculated from the sequence, used as the reference for the LC-MS identity check.
• Batch / lot number — the unique identifier for this specific production run. Critical: a COA only applies to the batch it names.
• Manufacture date and expiry / re-test date.

A COA without a clear issuing laboratory, batch number, or sequence should be treated with skepticism. These are the audit-trail anchors that let anyone re-verify the result.

The identity section should report the observed mass from LC-MS alongside the theoretical mass calculated from the declared sequence, with the deviation. A typical entry looks like:

Identity (LC-MS)
  Theoretical monoisotopic mass:   1419.5403 Da
  Observed monoisotopic mass:      1419.5391 Da
  Deviation:                       -0.0012 Da (-0.85 ppm)
  Result:                          PASS

Mass accuracy of better than ±0.1 Da on a high-resolution instrument is the standard threshold. The COA should also reference the instrument used (e.g., Orbitrap, Q-TOF) and the ionization mode.

Some COAs report identity as "confirmed by LC-MS" without numbers. That is weaker than reporting the actual masses, because it is unverifiable without the underlying data.

The purity section should report not just a number but the method:

Purity (RP-HPLC)
  Column:           C18, 4.6 x 150 mm, 5 micron
  Mobile phase:     A: 0.1% TFA in water; B: 0.1% TFA in MeCN
  Gradient:         5 to 60% B over 30 min, 1.0 mL/min
  Detection:        UV @ 214 nm
  Main peak area:   98.7%
  Largest impurity: 0.6% at RT 14.8 min
  Result:           PASS (spec >=95.0%)

A bare "≥99% pure" claim without the method conditions is uninformative — purity is a method-dependent number. Reproducibility requires knowing the column, mobile phase, gradient, and detection wavelength.

The largest impurity should also be reported, ideally with retention time and (if known) identity. A high purity number with an unidentified impurity at >0.5% is a flag worth investigating.

Recombinant-protein COAs (e.g. somatropin / rhGH) additionally report SEC-HPLC for dimer and aggregate content and HCP ELISA for residual host-cell protein — neither is measured by RP-HPLC at 214 nm. See the HGH testing methodology guide for how those two lines should read on a recombinant-protein certificate.

The content section answers the most-misunderstood question on a peptide COA: how many milligrams of active peptide are in the vial. Lyophilized peptides typically contain water (5 to 15%) and counter-ion (5 to 20%), so a vial labeled "5 mg of peptide" can easily contain 3.8 to 4.5 mg of actual peptide.

Content is reported either as:

• Amino acid analysis (AAA) — the gold standard. Reports the molar content of each amino acid after acid hydrolysis, allowing back-calculation of true peptide content.
• Quantitative HPLC vs. a calibrated reference standard — faster and less destructive, but only as good as the reference.
• UV absorbance at 280 nm — only valid for peptides containing Trp or Tyr; uses the calculated molar extinction coefficient.

A COA that omits content or reports content only as "labeled mass" is incomplete. The labeled mass is the dry weight, not the active peptide mass.

A complete release package also reports:

• Counter-ion identity and content — typically TFA or acetate, measured by ion chromatography. Reported as % of dry mass.
• Water content — by Karl Fischer titration. Reported as % of dry mass.
• Residual organic solvents — by headspace GC, against the limits in USP <467>.
• Heavy metals or elemental impurities — for clinical-grade material, against ICH Q3D limits.

Counter-ion content is the most commonly omitted — and the one most likely to materially change the active mass per vial.

Endotoxin testing is required for any peptide intended for parenteral (injectable) use. It is performed by a Limulus Amebocyte Lysate (LAL) assay or by recombinant Factor C, against the limit specified in USP <85> for the dose and route of administration.

A research-grade peptide intended for in-vitro use is not required to carry an endotoxin number, and the absence of one on the COA does not imply contamination — only that it was not measured. For any material that will enter a living system, look for an explicit endotoxin result with units (typically EU/mg) and a reference to the test method.

A COA should be signed by the responsible analyst or quality officer at the issuing laboratory. The document should reference the test methods (often by SOP number) and be reproducible from raw data files held by the laboratory.

A tamper-evident COA goes one step further: the document is content-hashed (SHA-256) and the hash is anchored to a public blockchain or another tamper-evident registry at issuance. Any later modification to the document would change the hash and break the anchor — which means anyone holding the COA can independently verify it has not been altered. This is the model CertikLabs uses for every verification page; see how the on-chain anchor works.

• No batch number, or a batch number reused across multiple production runs.
• Purity reported without the HPLC method (column, mobile phase, gradient, detection wavelength).
• "Confirmed by LC-MS" with no observed mass and no deviation from theoretical.
• No content / assay value — only the labeled dry mass.
• No counter-ion content.
• Endotoxin claimed without a numeric value and a method.
• An issuing laboratory that is the same entity as the manufacturer, with no independent verification available.
• No analyst signature or date of release.