Skip to main content
. Author manuscript; available in PMC: 2020 Jul 27.
Published in final edited form as: FEBS J. 2020 Apr 20;287(13):2685–2698. doi: 10.1111/febs.15314

Table 2. Evaluation of data imputation methods.

Mean results of 100 random experiments with standard deviations given in parentheses, in units of the last significant digit of the mean. MAD: Median Absolute Deviation, MAE: Mean Absolute Error, RMSE: Root-Mean-Square Error.

Error Method Clashscore RSRZ outliers [%] Ramachandran outliers [%] Rotamer outliers [%] R free [%]
MAD MICE 2.5(3) 2.01(2) 0.21(1) 1.23(3) 1.02(2)
Mean 4.4(3) 2.30(2) 0.49(1) 2.11(4) 2.50(4)
Median 2.9(3) 1.86(2) 0.04(1) 1.39(4) 2.50(4)
MAE MICE 3.7(4) 2.54(3) 0.41(2) 1.77(4) 1.31(2)
Mean 5.7(6) 2.74(3) 0.61(2) 2.56(6) 3.00(3)
Median 5.1(7) 2.60(3) 0.49(3) 2.34(7) 3.00(3)
RMSE MICE 5.7(15) 3.84(15) 0.84(11) 2.64(10) 1.77(3)
Mean 8.9(23) 4.04(14) 1.24(16) 3.65(14) 3.82(4)
Median 9.3(24) 4.17(14) 1.32(16) 3.85(15) 3.82(4)