Integrative Analysis Using Module-Guided Random Forests Reveals Correlated Genetic Factors Related to Mouse Weight
Figure 3
Summary of prediction errors of mgRF using three types of data.
(A) Boxplot of prediction errors in root-mean-square error (RMSE) in 10 trials of 10-folds cross-validation. Scatter plots of (B) genotype-only (x-axis) vs. expression-only (y-axis), (C) genotype-only vs. combined, and (D) expression-only vs. combined. The dashed diagonal lines (x = y) indicate points of equal RMSE. Given two vectors, the p-value of one-tail paired t-test evaluate if the distribution of one vector (100 RMSE values for 10 trials of 10-fold cross validation) is statistically smaller than the other one.