Fig. 4
From: An information-theoretic framework for deciphering pleiotropic and noisy biochemical signaling
Signaling dynamics has a potential to transfer information about identity and quantity of IFN-α and IFN-λ1. a Information capacity, \(C_A^ \ast\), as a function of the length of the output, tn, for different values of the differential kinetics coefficient, δ, and the coefficient of variation cv = 0. b Isolines of the quadratic forms (x − x0)TFIM(x0)(x − x0) for FIMs corresponding to different values of tn and δ. Given Eq. (11), these isolines show concentration changes that are detected with the same error. Further isolines have smaller error. Singularity (parallel lines) implies lack of discriminability. c Information capacity, \(C_A^ \ast\), as a function of the differential kinetics coefficient, δ for different values of cv assuming maximal considered output length, i.e., tn = 180. d As in b but for FIMs corresponding to c. Non-singularity of FIM indicates that both IFNs can be discriminated for sufficiently large number of cells. Modeling details: In b and d, we used x0 given by the receptors Kd’s, i.e., \(x_0 = (k_ - ^{\mathrm I}/k_ + ^{\mathrm I},k_ - ^{{\mathrm {III}}}/k_ + ^{{\mathrm {III}}})\). Remaining parameters used for computations are given in Supplementary Tables 1 and 2
