Skip to main content

Table 1 Inference results for (A) the simple model and (B) autoregulatory model of single gene expression

From: Bayesian inference of biochemical kinetic parameters using the linear noise approximation

(A)
Param.
Prior Value Estimate 1 Estimate 2
γ R Γ (0.44,10-2) 0.44 0.43 (0.27-0.60) 0.49 (0.38-0.61)
γ P Γ (0.52,10-2) 0.52 0.51 (0.35-0.67) 0.49 (0.38-0.61)
k P Exp(100) 10.00 21.09 (3.91-67.17) 11.41 (7.64-16.00)
λ Exp(100) 1.00 1.42 (0.60-2.57) 1.08 (0.76-1.36)
b 0 Exp(100) 15.00 6.80 (0.97-18.43) 12.78 (9.80-15.33)
b 1 Exp(1) 0.40 0.79 (0.05-3.02) 0.29 (0.18-0.43)
b 2 Exp(1) 0.40 0.77 (0.08-2.79) 0.77 (0.32-1.59)
b 3 Exp(10) 7.00 6.13 (4.41-7.85) 7.25 (6.79-7.55)
b 4 Exp(100) 3.00 0.94 (0.11-2.88) 2.87 (2.11-3.52)
(B)
Param.
Prior Value Estimate 1 Estimate 2
γ R Γ (0.44,10-2) 0.44 0.44 (0.27-0.60) 0.42 (0.32-0.54)
γ P Γ (0.52,10-2) 0.52 0.49 (0.33-0.65) 0.49 (0.36-0.61)
k P Exp(100) 10.00 14.86 (3.18-47.97) 9.35 (5.87-13.15)
λ Exp(100) 1.00 1.26 (0.48-2.30) 1.15 (0.81-1.50)
b 0 Exp(100) 15.00 5.99 (0.20-23.06) 13.47 (9.24-17.13)
b 1 Exp(1) 0.40 0.59 (0.01-2.75) 0.27 (0.14-0.53)
b 2 Exp(1) 0.40 0.92 (0.05-2.92) 0.83 (0.21-3.52)
b 3 Exp(10) 7.00 6.53(0.74-14.69) 7.27 (6.44-7.79)
b 4 Exp(100) 3.00 2.85 (0.35-7.19) 2.64 (1.82-3.32)
  1. Parameter values used in simulation, prior distribution, posterior medians and 95% credibility intervals. Estimate 1 corresponds to inference from single time series, Estimate 2 relates to estimates obtained from 20 independent time series. Data used for inference are plotted in Figure A for case A and Figure B for case B. Variance of the measurement error was assumed to be known σϵ = 9. Rates are per hour. Parameters are n H = 1, H = 61.98 in case B.