Figure 7

Summary of Significant Spots on Rabin Dataset: Multiple (20) datasets containing 10% missing values were generated by randomly removing spots values from the complete dataset containing 70 proteins. Missing values were then imputed using the Row Average method or KNN method with k = 5. Values in the Venn diagrams represent the number of discovered proteins (p-value < 0.05) in the original complete dataset and the imputed datasets. Note, to be discovered in KNN, RA, or LSM method for the imputed datasets, the spot needed to have a p-value less than 0.05 in at least half of the simulated datasets. The Venn diagrams refer to (a) parametric t test, (b) permutation t test, (c) Chebby Checker test, (d) normal-based bootstrap t test, (e) pivotal-based bootstrap t test, (f) percentile-based bootstrap t test.