Choosing a Peptide Testing Laboratory: Six Questions That Actually Matter

Laboratory comparisons in this market usually reduce to two variables: what it costs and how long it takes. Both are real considerations and neither has any bearing on what a result means.

The six questions below are the ones that decide whether a certificate can be used to conclude anything. They are listed in rough order of how much they change the answer.

1. What sampling model does the laboratory work under

This matters more than every analytical consideration combined, and it is almost never discussed.

A result is a statement about the material that entered the instrument. Extending it to a batch is an inference, and the strength of that inference is set entirely by how the sample was drawn. If one vial was submitted, the result describes one vial. If material was drawn across a defined production run, the result can support a statement about that run.

Neither model is a deficient version of the other. They answer different questions. What matters is knowing which question a given certificate answered, because a single-unit result presented as a batch statement is the most common overreach in this category. We have written the full argument in single vial versus batch representative testing.

Ask a laboratory what its minimum submission is and why. The answer tells you what its certificates are entitled to claim.

2. Does the certificate state the method

A purity figure is method-dependent. Change the gradient, the column, the run length or the detection wavelength and the same physical sample returns a different number, because impurities that co-eluted under one method separate under another.

The practical consequence is uncomfortable: a less demanding method reports higher purity. Nothing is falsified. The analysis simply could not resolve the difference. A laboratory optimizing for flattering numbers does not need to lie, it only needs to run a short gradient.

So the defense is disclosure. A certificate should state the technique, and ideally the detection wavelength. A number attributed to no method cannot be compared to any other number and cannot be reproduced by anyone. That is not automatically dishonest, but it removes every ability to check, which for a document whose entire purpose is verification is a strange thing to leave out.

3. Is identity confirmed, and shown

Purity establishes that the sample is predominantly one compound. It does not establish which compound. Confirming that requires mass spectrometry, and the certificate should report the measured molecular weight against the theoretical value for the claimed sequence.

Two failure modes are worth separating. A certificate that omits identity entirely has answered half the question. A certificate that states identity as confirmed without showing the measured mass has printed a conclusion, which is a different and slightly worse thing, because it looks complete.

Be equally clear about the ceiling. Molecular weight confirms mass, and mass is not sequence: the same amino acids in a different order produce the same weight. Routine commercial confirmation is mass plus chromatographic behavior against a reference standard, which is a sound standard when it is described accurately. A laboratory presenting that as sequencing is overselling, which tells you something about how it will describe everything else. See HPLC versus mass spectrometry.

4. Can a certificate be verified at source, by you

A certificate is a PDF, and PDFs are editable. The layout, the letterhead and the disclaimer language are the easiest parts to reproduce and the parts readers pattern-match on hardest.

What cannot be reproduced is a record held by the laboratory named on the document. So the question is whether the laboratory maintains one that a third party can query, and where that query resolves.

Detail that decides it: verification has to land on the laboratory's own domain. A QR code resolving to the supplier's website, a shared folder or any page the seller controls has verified nothing, because the party making the claim is hosting the confirmation of the claim. The correct procedure is to type the laboratory's domain yourself and enter the key manually.

Some genuine laboratories do not operate a public register. That is not misconduct, it just means their certificates are not verifiable by you and you are relying on the supplier's word that the document is theirs. Knowing which situation you are in is the point. More on this in how to verify a Certificate of Analysis.

5. Is the laboratory independent of the product

A laboratory that also manufactures or sells the compounds it certifies is testing its own work, and every result it issues carries that conflict whether or not it affects any individual measurement.

The related question is who commissioned the test, because sample selection sits upstream of the laboratory. When the party with an interest in the outcome also chose which unit to submit, the analysis is blind to a step that has already shaped the result. That is not a criticism of the laboratory, which will test what it receives accurately. It is a limit on what the certificate can be read to mean.

Novagen does not manufacture or sell compounds. We test them, and that is the entire business.

6. Does the laboratory describe its own limits

This one is a proxy, and it is a good one.

Honest certificates are narrow. Results pertain to the sample submitted. A Certificate of Analysis is not a product approval, not a safety certification, not an authorization for any regulatory purpose, and a chromatographic purity result is not a sterility or endotoxin result. That limiting language is a mark of a laboratory that understands what it produced.

Laboratories that drift toward endorsement, certifying a product as safe or vouching for a supplier, have confused two roles. Laboratories measure samples. They do not underwrite companies. How a laboratory talks about the boundaries of its own work predicts how it will talk about the results inside them.

Novagen and Janoshik: an honest comparison

This comparison gets searched for often enough that avoiding it would be conspicuous, so here it is stated plainly, including the part we cannot answer.

Janoshik Analytical is an independent analytical laboratory based in Slovakia and it is the name most people in this market encounter first. It is widely used for single-sample submissions, which is why it has the reach it does: a person holding one vial can send that vial and get a result on it.

Novagen Analytical Labs is an independent analytical laboratory in Cambridge, Massachusetts. We work with manufacturers and suppliers rather than individuals. We do not offer single-vial consumer testing. Testing engagements begin at a minimum of 1,000 units per submission, so that a result is representative of a batch rather than of one unit. Every certificate we issue carries a verification key that resolves on this site to the recorded result and the client it was issued to.

The meaningful difference is not quality, it is the question each model can answer.

• If you are holding a vial and want to know what is in that vial, a single-sample laboratory is the right tool and we are not. Send it to one.
• If you manufacture or supply a compound and want a certificate that can honestly describe a production run, batch-representative submission is what supports that claim, and it is what we are set up for.

Three things we will not tell you, because we cannot substantiate them.

We have not run split-sample comparisons against Janoshik or any other laboratory, so we make no claim about relative analytical accuracy. Anyone claiming a competitor's numbers are worse without published side-by-side data on identical material is asserting something they have not measured.

We do not hold ISO 17025 accreditation and we do not claim it. Accreditation is uncommon among laboratories serving this market generally, which means it is not a useful way to separate them and nobody should be presenting a badge here.

We are not a gold standard, and neither is anyone else. That phrase does no work. What can be checked is whether a laboratory states its methods, confirms identity, defines its sampling, and maintains a record you can query without going through the seller.

Things that look like criteria and are not

• Turnaround time. Speed is a scheduling property. It says nothing about method rigor and a fast result is not a better one. Ours is currently 21 days from receipt.
• Price. Cheap analysis and expensive analysis both produce a number. The difference shows up in method depth and review, neither of which is priced transparently anywhere.
• The look of the certificate. Design is the cheapest part of a document to get right and the first thing a fabrication copies. See spotting a fabricated certificate.
• A chromatogram image. A picture can be redrawn, reused or lifted from another report. Instrument data is informative when retrieved from the issuing laboratory and it is not evidence when accepted from a supplier's PDF.
• Reputation in forums. Useful signal about service and responsiveness. Not evidence about analytical method, which almost nobody posting has inspected.

The short version

Ask what the minimum submission is and why. Ask whether the method is printed on the certificate. Ask whether identity is confirmed by mass and shown. Ask where verification resolves. Ask whether the laboratory sells what it tests. Then read how it describes its own limits.

Six questions, all answerable in a short email, and a laboratory that answers them clearly has told you most of what a comparison could.

Next: how a submission actually works, or email admin@novagenlaboratories.com to discuss one.