Figure 3

Effects of the amounts of missing data on imputation procedures. 500 simulations were performed, where each simulation generated a datasets containing 5%, 10%, and 20% missing values by randomly removing spot values from the complete data set of 70 protein spots. Missing values were imputed by row average (Row Ave), LSM, KNN, and NIPALS imputation methods with k nearest neighbor values of 3, 5, or 8. Results of the imputation were compared using RMSE. The NIPALS methods are summarized by "nPR" which denotes the number of principal components used to impute the missing data.