Fig. 2: Analysis of Staurosporine dataset.
From: A Bayesian semi-parametric model for thermal proteome profiling
Condition A denotes the control and Condition B denotes 20 μM of Staurosporine (A) Melting profile for the DYRK1A with inferred mean sigmoid model function plotted, along with 95% credible bands for the inferred mean function. B Sensitivity for the different methods applied to the Staurosporine dataset. C Melting profile for AP4S1 using the sigmoid model, with uncertainty estimates in mean function. D Melting profile for AP4S1 using the semi-parametric model, including inferred mean function and 95% credible bands. E, F Posterior predictive checks for AP4S1 using the two Bayesian models: E sigmoid (F) semi-parametric. The red line corresponds to the observed data. Whilst the black line is the posterior predictive mean function and the credible bands correspond to 50% and 95% credible bands of the posterior predictive distribution, respectively. Statistics derived from two biological replicates, for each of two conditions each with 10 measure temperatures.
