A Practical Workflow for Antigen-Antibody Pair Screening

Updated 2026-09-28

A structured workflow for screening antibody pairs against a recombinant antigen, from coating controls to sandwich confirmation in the intended matrix.

Define the Assay Before Screening

Pair screening works best when the assay format is defined first. Decide the sample type, the required working range and the detection chemistry before testing antibodies, because these choices constrain which pairs can succeed. Establish a reference antigen preparation and keep it constant across the screen, so that differences in signal can be attributed to the antibodies rather than to the antigen. Changing the antigen mid-screen makes the resulting comparison meaningless. Plate the antigen at a fixed concentration and reserve it as a control across every plate in the run.

Stepwise Screening

Screen individual antibodies first as capture, using the reference antigen and a generic detection system, to rank binding and background. Then test candidate pairs in a sandwich format with the antigen in the intended matrix, applying each pair in both orientations because the results often differ. Include a no-antigen blank and a negative control on every plate. Rank pairs by signal-to-noise, dose response and consistency across replicates rather than by absolute signal, which varies with coating and detection conditions.

Confirming a Working Pair

A pair that performs in buffer should be re-tested in the intended sample matrix, where proteins and interfering substances can change the result. Check dilution linearity and recovery across the range, and confirm that the pair tolerates the sample handling you plan to specify. Examine performance at the ends of the curve, since a pair can appear sound at mid-range while failing at low concentrations. Record the orientation, the concentrations and the matrix for every pair you retain, so the selection can be reproduced.

Reading a failed pair: which component is actually at fault

When a candidate pair underperforms, the useful next step is to identify which component is responsible rather than to discard both. Capture and detection antibodies do different jobs and fail in different ways. A capture antibody that performs well alone but contributes nothing in a sandwich may be binding an epitope that the detection antibody blocks, which is a pairing problem rather than a defect. A detection antibody that produces high background in buffer but acceptable signal in matrix is usually a labelling or blocking issue rather than a specificity issue. The cheapest diagnostic is to exchange one component at a time against a known-good partner, keeping everything else constant — which is only possible if a known-good partner exists in your laboratory. If it does not, the first experiment of a screening programme is to establish one.

Recording the screen so it can be repeated

Pair screening produces large amounts of comparative data that is difficult to reconstruct later, so what you record determines whether the work is reusable. For every pair retained or rejected, record the orientation tested, the coating concentration, the detection concentration and dilution buffer, the antigen concentration and its lot, the matrix, and the incubation timing. Note the blanks and controls on every plate, since a comparison across plates is meaningless without them. Record the result as a ratio or a normalised signal rather than as raw absorbance, because plate-to-plate variation will otherwise dominate a later re-reading of the data. And record the reason for rejection in words, because 'poor' is not reproducible information — 'high background in buffer, low signal at the bottom of the range, orientation B only' is.

Ask a technical question