Fig. 9: A simulation of the impact of K_D optimization using TEA versus inverse-Langmuir.

a Target concentration T(t) used in this simulation. b Simulated response of two receptors with fixed \({{{{{{\rm{k}}}}}}}_{{{{{{\rm{on}}}}}}}={10}^{6}\) s⁻¹ M⁻¹ but different K_Ds to a spike in target concentration. Also shown for comparison is the noise N[n] introduced by the detector system (black dots). In this simulation, N[n] is zero-mean white noise with a standard deviation of 0.005 fraction bound. c The response of the receptors was sampled at f_S = 1/15 Hz and the noisy samples were digitally low-pass filtered with f_T = 1/400 Hz. The data was then used to reconstruct the original target concentration either using the pre-equilibrium TEA (solid lines) or conventional inverse-Langmuir (dashed lines) for the receptor with K_D = 1 nM (red) or 10 nM (orange). d We calculated the error in the target concentration after repeating the simulation with receptors with k_on = 10⁶ s⁻¹ M⁻¹ but a range of K_Ds. This is shown in terms of SNR for both the pre-equilibrium TEA (solid line) and the simple inverse-Langmuir (dashed line). Vertical red and orange lines correspond to the simulations in (b) and (c).