Fig. 2From: Identification of in vitro and in vivo disconnects using transcriptomic dataFractional polynomial framework example: Using gene A2m and compound sulindac. Illustration of changes in predicted profiles by fitting fractional polynomial models with different powers on the same gene expression data. The model in the bottom right panel with p 1,p 2=2 is the best predictive model with the minimum Akaike’s Information Criterion (AIC)Back to article page