Figure 1

Simulation results showing the capability of GLFD to discover latent factors, as compared with PLS and SPC. In every simulation, 5 modules, each consisting of 200 simulated genes, were generated. The first module was governed by the clinical factor, together with 1~3 other latent factors (columns). The other four modules were governed by 2~4 factors. All factor scores were drawn independently from the standard normal distribution. Gaussian random noise was added to achieve different signal to noise ratios (rows). An additional 1000 pure noise genes were generated from the standard normal distribution. Each simulation setting was repeated 100 times. The success of latent factor recovery was evaluated by the R² values obtained by the regression of each latent factor against the identified factors. The relative frequencies (10 equal-sized bins between 0 and 1, equivalent to the histogram) of the R² values are plotted.