Fake Peptide COA: How to Spot Forged & Misleading Reports

A certificate of analysis is the only evidence most people ever see about what is inside a research vial. That makes it the most valuable document in this field, and the most valuable one to fake — a COA is a PDF, trivially editable, and a plausible chromatogram can be pasted into a template by someone with no laboratory involved at all.

Fabrication leaves fingerprints, though, and the chemistry under a report is harder to fake than the layout. Read a COA as a document examiner and an analytical chemist would together: is it real, is it about the material in front of you, and does it say what someone claims? You need a PDF reader, a calculator, a chemical database, and the willingness to email a laboratory.

Research use only. These are research chemicals; most are not approved for human use. This page is document literacy — how to read and verify a laboratory report — not health guidance. It contains no dosing, administration or medical advice.

The three failure modes, kept separate

Nearly every COA problem is one of three things. They need different checks and carry different weight, so keep them apart.

1. Outright forgery

The document was fabricated, or a genuine one edited — a purity figure raised, a compound name swapped, a lot code overwritten, a logo pasted onto a template. The rarest of the three, and the most complete failure: nothing on the page is evidence of anything.

2. Genuine document, misrepresented

Far more common. The lab is real and the numbers are real, but the document does not describe the material under discussion: a report for a different lot, an old report presented as current, a supplier’s in-house COA passed off as independent testing, or someone else’s report screenshotted from a forum. The claim attached to it fails, not the document.

3. Technically true, materially misleading

Every number is accurate and the impression is still wrong. One favourable lot implied to cover all production. Purity with no identity test, so nothing establishes that the pure substance is the right substance. An area percentage read as how much peptide is in the vial. Sterility or endotoxin implied by a document that tested neither.

Key insight: Usually the problem is not a fake report but a real one doing work it was never designed to do. Ask what was tested, on what sample, on what date — before asking whether it is genuine.

Anatomy of a COA

A complete report is a chain of custody in miniature: a sample, a method, a result, a date, a person. A gap anywhere breaks the link between paper and powder.

Field What a complete report shows What its absence means
Issuing laboratory Name and contact details Unverifiable by construction
Report / verification ID A code tying the PDF to a record at the lab No handle for verification
Sample name and lot code The compound as declared, plus a batch code also on the vial Attaches to no particular material
Date of analysis At or after the lot’s manufacture The result may predate the material
Method Column, mobile phase, gradient, wavelength, MS mode Cannot be judged or reproduced
Chromatogram and impurity table Labelled axes, baseline noise, integration marks, peak areas A results table alone is an assertion
Analyst and signature A named person and an issue date Nobody owns the numbers

Document-level red flags

Before any chemistry, look for the gaps a forger cannot fill.

  • No lot number, or one that does not match the vial. The report attaches to nothing, or to other material.
  • No test date, or a date preceding the lot. A report cannot analyse material that did not yet exist.
  • No method, or a results table with no chromatogram. The chromatogram is the evidence; the table only summarises it.
  • No analyst signature, and cropped screenshots excluding the header, date or verification code.

Image and PDF forensics you can actually do

Typography, alignment and text layers

In a lab-generated PDF the numbers and method text are usually selectable; a report that is one flat image has lost its structural evidence. Zoom to 400% and compare a suspect figure with the same digits elsewhere: different font weight, different digit spacing, a baseline sitting fractionally high or low, or a faint halo and blocky artefacts around a retouched number where compression was applied twice.

Read the PDF properties

Most readers expose these — File → Properties (Cmd/Ctrl-D) in Acrobat, Tools → Show Inspector (Cmd-I) in macOS Preview — giving Creator, Producer and the creation and modification dates. Two patterns deserve a question: a Producer that is a photo editor rather than a report generator, and modification long after creation. Neither proves tampering, since legitimate documents get flattened and re-exported constantly.

Compare chromatograms across reports

The strongest check available to a layperson. Baselines carry random detector noise, and no two injections produce identical noise — not on the same instrument, not on the same sample. Peaks can legitimately look alike; baseline wobble cannot match wiggle for wiggle, so compare the flat regions, not the peaks.

Red flag: Identical chromatogram traces under different lot numbers or different products. One reused image discredits the chromatographic evidence in every report it appears in.

Numbers that are too clean

Purity landing on exactly 99.9% across every product and lot, or quoted to more decimals than the method supports, deserves a look. So does an impurity table that will not reconcile: main peak plus listed impurities should account for essentially 100%, allowing for rounding and any declared threshold.

Chemistry-level tells

Does the observed mass match the real molecule?

Look the compound up in a public database such as PubChem — BPC-157, for instance, is C62H98N16O22, about 1419.5 Da. Then read the mass spectrometry section carefully, because the most common false alarm in COA reading happens here.

Electrospray ionisation does not give one peak at the molecular weight. It gives multiply protonated ions plotted as mass-to-charge, so the ion carrying n protons appears at m/z = (M + n × 1.007)/n. For a 1419.5 Da peptide that is roughly 1420.5 singly charged, 710.8 doubly, 474.2 triply — so a report showing 710.8 is correct. Others give a deconvoluted neutral mass, back-calculated from the charge series, which should sit near the molecular weight itself.

Two more legitimate discrepancies, so you do not raise false flags. Average and monoisotopic masses differ by a dalton or more at a few thousand daltons; and MS sees the peptide, not its acetate or trifluoroacetate counterion, so a mass matching the free peptide on a product labelled as an acetate salt is right. Worry only about a mass irreconcilable under any convention.

Does the purity match the picture, and the method the analyte?

Peptide purity is normally calculated by area normalisation: main peak area over total integrated area, with UV detection around 210–220 nm where the peptide bond absorbs. So you can check it by eye: a shoulder or second peak that is visibly a meaningful share of the main peak’s area is not consistent with a stated 99.8%. The workhorse method is reversed-phase HPLC on a C18 column with a water/acetonitrile gradient and about 0.1% trifluoroacetic acid, identity by ESI-MS. Retention time is weaker than people assume — meaningful only against a stated method, so comparing across labs proves nothing.

The purity trap: area percent is not how much peptide is in the vial

This is the most exploited gap between a true statement and a false impression. A purity of 99% by area normalisation means that of the UV-absorbing material which eluted and was integrated, 99% of the area was the main peak. It says nothing about the mass of powder in the vial.

Lyophilised peptides also contain water and a counterion — trifluoroacetate or acetate, formed during synthesis and purification — plus residual solvent and salts, none visible to a UV detector at 214 nm. Manufacturer QC literature commonly places net peptide content for research-grade material well below gross weight — figures around 60–80% are typical, and the exact value is sequence-dependent, since peptides rich in basic residues bind more counterion. So 99% pure and roughly three-quarters of this powder is peptide can both be true of one vial.

Different questions, different methods: purity by RP-HPLC area normalisation; water by Karl Fischer titration, which quantifies water through its stoichiometric reaction with iodine and sulfur dioxide; counterion by ion chromatography, HPLC or gas chromatography; net peptide content by amino acid or elemental analysis. A COA reporting only area-percent purity has not measured content.

Key insight: Purity, identity, quantity and content are four different tests. Purity says the main peak dominates; identity says the mass matches the target; quantity says how many milligrams are present; content says what fraction of that mass is peptide rather than water and salt. A document answering one is not evidence about the others.

An honest caveat: there is no universal purity spec for research peptides. ICH Q3A, which sets impurity thresholds for new drug substances, explicitly excludes peptides from its scope, and chapters such as USP <621> govern system suitability rather than dictating an acceptance number. Any hard industry-wide cut-off is a convention, not a standard.

The step that matters most: verify with the laboratory

Everything above narrows the field. One action closes it: confirming the report with the lab that issued it. Whoever presents a COA has an interest in it; the laboratory does not.

Several analytical laboratories used in this space issue reports carrying unique identifiers for exactly this purpose. Janoshik Analytical, for example, prints a task number and a unique key, and hosts a verification page on its own site where that pair can be checked. Others serving this market — MZ Biolabs and Colmaric Analyticals among them — publish their methods and contact details. Naming them is factual, not an endorsement.

Reach the lab’s site by typing the address yourself rather than following a link supplied with the COA, since a link is as forgeable as the document — and check that the record matches on compound, lot, date and values, not merely that something came back.

Where no portal exists, email the lab: give the report number, the date and the sample name as printed, attach the PDF, and ask whether that report exists and matches. Some labs confirm only that the number corresponds to what you sent — still enough to separate forged documents from genuine ones.

Red flag: A COA that cannot be confirmed with the issuing lab — no report ID, no contact, no reply — is unverified. Unverified is not the same as fake, and it is not the same as verified. Treat it as an open question, not as evidence.

The red-flag checklist

  • No lot number, or one absent from the vial or label
  • No analysis date, or a date preceding the lot
  • No laboratory name, contact details or report identifier
  • No method, no chromatogram, or a chromatogram with no axes or integration marks
  • No analyst signature
  • Low-resolution raster chromatogram inside an otherwise crisp vector PDF
  • Identical baseline noise across different lots or products
  • Fonts, digit spacing or row alignment inconsistent around one number, or metadata showing an image editor as Producer
  • Purity always exactly 99.9%, or more decimals than the method supports
  • Reported mass irreconcilable with the molecular weight under any charge state
  • One lot’s result used to characterise all production

What a good verification workflow looks like

  1. Establish what was tested. Read the sample name and lot code before any result; missing either, stop.
  2. Match the lot to the physical item — the report’s code against the vial’s.
  3. Check the dates — analysis at or after manufacture, and recent enough to matter.
  4. Read the method. Confirm it exists, suits a peptide, and names a wavelength and MS mode.
  5. Interrogate the document. Select the text, open the PDF properties, zoom to 400% on key figures, compare baselines against other reports.
  6. Do the arithmetic. Reconcile the observed mass through the charge states; check purity against the peaks and the impurity table.
  7. Verify with the issuing lab — its verification page, reached by typing the address yourself, or email.
  8. Record the outcome as verified, unverified or contradicted, and note which of the four questions the document answered.

What even a verified COA does not tell you

A report that survives every check above is strong evidence about one thing: a specific sample, drawn at a specific moment, analysed by specific methods. It is not a property of the vial in someone’s hand.

The gaps are structural, not accusatory. Someone selected and submitted the sample, and a lab reports only on what it received; a lot is not homogeneous by assumption, so one vial’s result is an inference about the rest. Lyophilised peptides are sensitive to moisture, heat and time, so a report issued before shipping says nothing about what shipping did. And no method can certify that the vial received came from the lot described.

The only way to close the gap is testing one’s own sample at a laboratory engaged directly, with a chain of custody one controls; some labs accept submissions from individual researchers. That result describes the material actually in hand — a property no supplied COA can have, however genuine.

Does a verifiable report number prove the peptide is good?

No. It proves the document is genuine and matches a record at that lab. Whether the values describe the material in question rests separately on the lot code matching and the vial coming from that lot. Verification defeats forgery; it does not defeat misrepresentation.

The mass on the COA is about half the molecular weight. Is it fake?

Almost certainly not — that is normal for electrospray ionisation, which produces multiply protonated ions plotted as mass-to-charge. A doubly charged ion sits near (M + 2 × 1.007)/2, close to half the molecular weight; a triply charged one near a third. Check whether the figure fits a charge state first.

Why can a COA say 99% pure when much of the powder is not peptide?

Because purity by area normalisation is a ratio of peak areas among UV-absorbing material, not a mass fraction of the vial. Water, counterion and salts are largely invisible at around 214 nm, and are measured by Karl Fischer titration, ion chromatography and amino acid analysis — tests a purity-only report has not performed.

What purity number should a research peptide report show?

There is no authoritative threshold. ICH Q3A explicitly excludes peptides from its scope, and pharmacopoeial chromatography chapters address system suitability rather than acceptance criteria. Any figure quoted as an industry standard is a convention; what is checkable is whether the purity reconciles with the chromatogram and impurity table.

Sources & Further Reading