MACS Matchmaker
Should you wait for equilibrium, or fit kinetics? This tool answers that question for a given binding pair (kon, koff) and titration series, and shows what happens to your apparent KD if you treat a non-equilibrated endpoint as if it were steady state.
Reaching steady state takes ~3/kobs. For tight binders that can be hours per concentration. The verdict tells you whether to wait for equilibrium or fit kinetics; the isotherm shows what the apparent KD looks like if you fit the endpoint anyway.
kon (M⁻¹ s⁻¹)
koff (s⁻¹)
Endpoint experiment
Injection duration (s)
KD = koff / kon = 10 nM
Lowest C (320 pM): kobs = 1.0e-3 s⁻¹ · t95 = 48 min
Highest C (320 nM): kobs = 3.3e-2 s⁻¹ · t95 = 1.5 min
Verdict is set by the lowest C — the slowest curve to equilibrate.Apparent KD from 2 min endpoints
True KD: 10 nM
Apparent KD (fit): 95 nM
Bias: 9.5× weaker (+850 %)
Apparent fit also recovers Rmax ≈ 120 pg/mm² (true 100 pg/mm²).Equilibrium reached at injection end
Lowest C (320 pM): 12 % of Req
Highest C (320 nM): 98 % of Req
The slowest curve to equilibrate is always at the lowest C (kobs is dominated by koff there). The bias in apparent KD is set by the slowest curve, not the fastest.Sensorgram — vertical line marks the injection end. The lowest C is slowest to equilibrate and sets the bias.
Isotherm — true Langmuir vs. fit through the endpoints
Caveats
- Pseudo-first-order assumed (analyte concentration constant during the injection).
- No mass-transport limitation, no rebinding, single-site 1:1 binding.
- Endpoint analysis is fine for fast off-rates; for slow off rates, prefer kinetic fitting over endpoint KD.
- Measured response noise and baseline drift are not modelled here — they make endpoint analysis worse, not better.