Skip to main content

Table 1 Values for the range of RMSE and range of Rank Change for each of 7 imputation methods, separated by the number of assays per slice

From: Characterizing the effects of missing data and evaluating imputation methods for chemical prioritization applications using ToxPi

Slices/Assay  
Method Measure 1 5 10 15 20 30 50 90 125
Binomial Rank Change (-210,-40) (-150,2) (-74,4.2) (-37,4.5) (-26,4.6) (-15,4.6) (-8.8,4.6) (3.2,4.6) (3.1,4.6)
Binomial RMSE (0.024,0.35) (0.058,0.25) (0.083,0.27) (0.086,0.24) (0.12,0.26) (0.12,0.25) (0.13,0.24) (0.15,0.24) (0.16,0.24)
kNN Rank Change (-210,-57) (-210,-59) (-210,-55) (-210,-55) (-210,-62) (-210,-61) (-210,-57) (-210,-48) (-210,-45)
kNN RMSE (0.016,0.16) (0.017,0.23) (0.017,0.22) (0.019,0.2) (0.018,0.22) (0.018,0.25) (0.019,0.22) (0.019,0.21) (0.017,0.2)
LLS Rank Change (-190,-12) (-56,-1.7) (-27,-0.31) (-25,1) (-19,1.4) (-14,1.6) (-11,2) (-8.7,2) (-6.6,2)
LLS RMSE (0.019,0.23) (0.043,0.43) (0.077,0.41) (0.09,0.45) (0.054,0.42) (0.1,0.45) (0.13,0.31) (0.14,0.24) (0.14,0.23)
Maximum Rank Change (-54,-1.7) (-210,-11) (-29,0.53) (-22,0.88) (-18,1.4) (-14,1.8)   (-11,2) (-8.3,2) (-6.4,2)
Maximum RMSE (0.031,0.78) (0.36,0.76) (0.45,0.68) (0.46,0.67) (0.5,0.66) (0.54,0.65) (0.55,0.64) (0.57,0.63) (0.57,0.64)
Mean Rank Change (-210,-11) (-54,-1.7) (-29,0.53) (-22,0.88) (-18,1.4) (-14,1.8) (-11,2) (-8.3,2) (-6.4,2)
Mean RMSE (0.019,0.26) (0.043,0.23) (0.069,0.28) (0.08,0.24) (0.097,0.26) (0.1,0.24) (0.12,0.24) (0.14,0.23) (0.15,0.23)
Minimum Rank Change (-210,-57) (-180,-21) (-25,0) (-110,-7.5) (-91,-4.8) (-73,-1.9) (-50,-0.76) (-31,0) (-140,-9.2)
Minimum RMSE (0.014,0.14) (0.031,0.16) (0.047,0.15) (0.065,0.16) (0.074,0.16) (0.086,0.17) (0.093,0.17) (0.11,0.17) (0.11,0.17)
SVD Rank Change (-210,-49) (-160,-17) (-130,-5.5) (-97,-4.4) (-81,-3.4) (-64,-1.2) (-51,-0.15) (-34,0.15) (-25,0.15)
SVD RMSE (0.02,0.46) (0.04,0.62) (0.073,0.54) (0.097,0.54) (0.096,0.57) (0.11,0.54) (0.1,0.54) (0.12,0.53) (0.11,0.53)