Dilution Is a Hypothesis, Not a Procedure: Rethinking Antibody Titration in UK Laboratories
A datasheet dilution has a seductive quality: it is precise, printable and easy to put in a protocol. That is also its danger. Antibody binding is contextual, and a dilution optimised by a supplier in one buffer, with one lot, on one platform, is not a local operating procedure. In UK laboratories, where sample matrices vary, staff rotate through shared facilities and automation is increasingly the norm, titration deserves to be put back on the agenda as a routine act of evidence generation.
The datasheet is a range, not a rule
Suppliers usually provide a recommended dilution range because they cannot know your antigen density, buffer composition, detection chemistry, incubation time or instrument. A datasheet dilution is best read as a hypothesis: a reasonable place to begin. It becomes a procedure only after local verification.
That distinction matters more than it once did. UK diagnostic and research laboratories are under pressure to do more with less, and copying a dilution from a paper or a previous project can look like efficiency. In practice, it transfers unexamined assumptions. A dilution that works in a clean recombinant antigen ELISA may fail in serum, plasma, tissue homogenate or a complex environmental swab matrix. A dilution that performs on a manual plate may behave differently on a locked automated platform with low reagent volumes and fixed wash steps.
Where dilutions fail in practice
Matrix effects are the first pressure point. Biological matrices can mask epitopes, introduce interfering antibodies, alter ionic strength or compete for binding. In UK clinical and veterinary diagnostic work, the same antibody may be asked to perform in several matrices across different services. A single dilution cannot be assumed to cover them all.
Lot-to-lot drift is the second. Antibody production is a biological process, and even well-controlled suppliers can show variation in purity, aggregation, conjugation ratio or functional activity. A laboratory that has not established its own performance envelope may find that a new lot shifts the signal-to-noise ratio just enough to affect interpretation.
Automated platforms add a third layer. Smaller reagent volumes, different incubation temperatures and platform-specific wash efficiency can change the effective working concentration. Shared core facilities add a fourth: multiple users, different buffers, different storage practices and no single owner of the reagent's performance history. In that environment, a dilution copied from a colleague's notebook is not a controlled document.
A simple titration framework for 2026
A practical framework does not need to be elaborate. It needs to be repeatable, documented and linked to intended use. Six steps would make a sensible minimum.
- Define the intended use and matrix. State whether the antibody is for research use, assay development or a diagnostic workflow, and identify the matrices in scope.
- Fix the variables you can. Use the same antigen batch, buffer, detection system and instrument for the titration series, and record them.
- Run a dilution series around the expected working range. A five-point curve in triplicate is often enough to see the shape of the response; a checkerboard against antigen concentration is better where the antigen load is uncertain.
- Include negative matrix and blank controls. Signal-to-noise, not signal alone, should drive the decision.
- Set acceptance criteria before the experiment. Define acceptable background, dynamic range, precision and, where relevant, limit of detection.
- Record a working dilution with a guard band. The chosen dilution should sit in a robust part of the curve, not on a cliff edge, and should be re-verified on lot change, instrument change or conjugation change.
For laboratories working under quality systems, this evidence can support method verification and change control. For IVD developers, it feeds directly into the technical file and UKCA readiness. For research groups, it is simply good practice: a titration record is cheaper than repeating an experiment because a new lot behaved unexpectedly.
Governance, not just pipetting
Titration is often treated as a bench task. It is also a governance task. Shared facilities need a named reagent owner, a controlled SOP, a lot register and a change-control route. Electronic laboratory notebooks and LIMS can make this easier, but the cultural point matters more: a working dilution should be treated as a living document, not a line in a method section that nobody revisits.
An annual review is a reasonable discipline. Has the antibody been re-titrated since the last lot change? Has the platform changed? Have new sample matrices been introduced? If the answer to any of these is yes, the old dilution is history, not evidence.
The dilution delusion is not that datasheets are wrong. It is that they are sufficient. In UK laboratories in 2026, the better habit is to treat every working dilution as a local claim that needs local support. That is not bureaucracy. It is the difference between an assay that happens to work and an assay that can be shown to work.
