Sample Dilution Range Calculation for Competitive ELISA kits

A researcher reached out before starting a new batch of competitive ELISA runs. They were measuring prostaglandin E2 (PGE2) in bovine plasma to compare an inflammatory response between two treatment groups, and had received conflicting advice from a colleague about whether their previous plates had been run correctly. Two questions needed answering: did the data look like it came from a correctly run assay, and could they keep running undiluted plasma when every other kit in the project had required dilution? They shared their standard curve, duplicate wells and interpolated sample values for review.

Because this is a competitive ELISA, optical density falls as analyte concentration rises — the opposite of a sandwich assay. A healthy curve therefore has its highest OD at the blank and its lowest at the top standard.

Values are illustrative and rounded to show the pattern, not the researcher's exact figures.

Two things trip people up on competitive assays. First, the inverted curve: a blank that reads lower than a standard looks "wrong" only until you remember that in a competitive format the zero standard should sit at the top. When a single blank replicate reads below its neighbours, it is almost always a pipetting or bubble artefact rather than a failed plate. Second, the dilution question is really a range question. Other assays in a project may need dilution because their targets are abundant and push samples above the top standard; PGE2 in plasma frequently sits in the tens-to-low-hundreds pg/mL, landing mid-curve without any dilution at all.

The standard curves were consistent and correctly shaped, and re-fitting one of them with a 4-parameter logistic (4PL) regression reproduced the researcher's interpolated concentrations almost exactly — a good sign the original analysis was sound. Both sample groups landed in the reliable middle of the curve, where a small change in OD maps to a proportionate, trustworthy change in concentration. The one value to set aside was a single blank replicate reading below the rest: since the blank should be the highest point on a competitive plate, that lone low reading is an outlier, not a curve failure. A few duplicate pairs showed wider OD spreads than ideal, which is worth tightening but does not invalidate the run.

Why is my blank higher than my standards?

That is exactly what a competitive ELISA should do. With no analyte to compete for antibody, the zero standard develops the most signal, so its OD is the highest on the plate and OD falls as concentration climbs.

Do I need to dilute my plasma samples?

Not if their interpolated concentrations already fall within the kit's 15.6–1000 pg/mL range and sit toward the middle of the curve. Dilute only when a sample reads above the top standard.

Which curve fit should I use?

A 4-parameter logistic (4PL) regression. It models the sigmoidal shape of a competitive curve accurately across the whole range, unlike linear or point-to-point fits.

What CV should I expect between duplicates?

Aim for 20% or lower. Higher variation usually points to pipetting inconsistency, bubbles, wells drying out, or temperature drift during incubation.

Bovine Prostaglandin E2 (PGE2) ELISA Kit (BOEB1203) →

SKU	BOEB1203
Detection range	15.6–1000 pg/mL
Sensitivity	8.1 pg/mL
Sample types	Serum, plasma, cell culture supernatants, tissue homogenates
Format	Competitive ELISA, 96 assays
Price	$857 — confirm current price before publishing

View product →

For research use only. Not for use in diagnostic procedures. Case details have been anonymized and sample values are illustrative; assay performance depends on your samples, protocol and instrumentation.