Affilabs.core SPR data analysis software

SPR data analysis follows a consistent order: inspect the raw sensorgrams, subtract the reference channel, choose steady-state or kinetic fitting, fit the simplest model (1:1 Langmuir) first, then judge the fit by its residuals and the plausibility of the constants. Common analysis errors come from skipping a step or forcing a complex model onto flawed data.

1. How do you read SPR data?

   An SPR sensorgram plots response (RU) against time through five phases: baseline, association, steady state, dissociation and regeneration. The rising association curve reflects binding, its plateau approaches equilibrium, and the decay after the analyte is replaced by buffer reflects dissociation. Before fitting anything, check that the baseline is stable and the curve shapes look plausible.

   Walk through each phase in SPR sensorgram explained, and match odd curve shapes to their causes with the sensorgram troubleshooting cheat sheet.

   In Affilabs.core: a live 4-channel sensorgram with automatic injection flags and cycle markers, plus real-time signal quality scoring with leak and bubble detection. Sparq, the built-in assistant, also flags common data patterns such as mass transport, bulk refractive-index shifts, drift and incomplete regeneration.

2. Why subtract a reference channel in SPR?

   Reference subtraction removes signal that is not specific binding. A reference channel carries the same surface without the ligand, so subtracting it cancels bulk refractive-index shifts from the sample buffer, temperature drift and much of the non-specific binding. Many labs also subtract a buffer-only (blank) injection, known as double referencing, to remove remaining systematic artifacts.

   In Affilabs.core: set the reference channel in the Edits tab. On the P4SPR 2.0, four channels run together — three active and one reference. Double referencing (also subtracting a blank buffer injection) is supported.

3. Should you fit SPR data with a steady-state affinity or a kinetic model?

   Fit steady-state affinity when you only need K_D: plot the equilibrium response against analyte concentration and fit a binding isotherm for K_D and R_max. Fit kinetics when you need on- and off-rates: fit the association and dissociation curves globally for k_a and k_d, then K_D = k_d / k_a. Kinetic data need controlled, continuous flow.

   Steady-state fitting is only as good as the plateau: each concentration must be at, or very close to, equilibrium when you read the response.

   In Affilabs.core: steady-state and kinetic fitting with the basic models: a 1:1 Langmuir isotherm for steady-state K_D, and a 1:1 kinetic fit for k_a, k_d and K_D on flow data. The P4SPR 2.0 is a static (steady-state) instrument; the P4PRO with AffiPump adds controlled flow for kinetics. See the Starter Test Kit protocol for a K_D fit step by step.

4. Why is the 1:1 Langmuir model the default SPR fitting model?

   The 1:1 Langmuir model assumes one analyte molecule binds one independent, identical site, which is the simplest possible binding mechanism. It has the fewest fitted parameters, so its constants are the most robust. Start with it, and treat systematic deviations from it as a clue about your assay before reaching for a more complex model.

   In Affilabs.core: select the Langmuir 1:1 model in the Binding subtab and click Calculate K_D; the fit reports K_D, R_max and R².

5. When should you use a heterogeneous or two-state SPR fitting model?

   Only when independent evidence supports the mechanism. Heterogeneous-ligand, two-state (conformational change) and bivalent-analyte models add parameters, so they almost always fit better than 1:1, even when the mechanism is wrong. Biphasic curves often come from a mixed or overloaded surface, aggregation or mass transport. Fix the surface, ligand density and concentration range first.

   Supporting evidence for a complex model includes a known second binding site, a documented conformational change, or the same deviation persisting across ligand densities and surface chemistries. More on this in measuring K_D for protein–protein interactions.

6. How do you know if an SPR curve fit is good?

   Look at the residuals, the difference between data and fit. They should scatter randomly around zero at the noise level; systematic waves mean the model does not describe the data. Also check that fitted values are physically plausible: R_max consistent with ligand density, and a concentration series spanning 0.1× to 10× K_D. A high R² alone is not proof.

   In Affilabs.core: residual plots are shown with the fit, R² is reported with the Langmuir 1:1 K_D fit, and data export to Excel or CSV for Origin and Prism. For advanced models (heterogeneous ligand, two-state, mass transport), export the data to a specialized fitting tool.

7. What is mass transport limitation in SPR?

   Mass transport limitation occurs when analyte reaches the sensor surface more slowly than it binds, so delivery rather than the interaction controls the signal. Signs include a nearly linear association phase and rate constants that change with flow rate. Reduce ligand density, raise the flow rate, and test two or three flow rates before trusting kinetic constants.

   Fitting models with a mass-transport term exist, but they correct for the effect rather than remove it; a well-designed experiment is the better fix.

8. What is the difference between single-cycle and multi-cycle kinetics?

   Multi-cycle kinetics injects one analyte concentration per cycle, with regeneration between cycles, giving one complete sensorgram per concentration. Single-cycle kinetics injects increasing concentrations one after another in a single cycle, with a long dissociation phase at the end and no regeneration in between. Single-cycle suits ligands that are hard to regenerate; both are fitted globally.

   For a side-by-side of manual injection and pump-assisted runs, see Manual injection vs pump-assisted SPR.

   In Affilabs.core: the Method Queue builds multi-step run sequences from templates, and steps can be added mid-run. Templates cover both single-cycle and multi-cycle kinetics.

Further reading: SPR sensorgram explained · SPR binding affinity for protein–protein interactions · Sensorgram troubleshooting cheat sheet