Table 4

The 16 mis-classified samples

A_id
Animal
Treatment
Cpd.Dose.Day
H_score
B_class
SVM_value
Class_predicted
TRUE/FALSE

84
2021
Tobramycin
Tob.060.03
0
-1
0.053417599
1
FALSE
80
2017
Tobramycin
Tob.030.14
0
-1
0.020981321
1
FALSE
36
2071
HCB
HCB.040.03
0
-1
0.3853432
1
FALSE
5
2117
Allopurinol
All.030.03
0
-1
0.4184979
1
FALSE
6
2118
Allopurinol
All.030.03
0
-1
0.33539873
1
FALSE
7
2119
Allopurinol
All.030.03
0
-1
0.079393616
1
FALSE
8
2120
Allopurinol
All.030.03
0
-1
0.40372499
1
FALSE
17
2142
Allopurinol
All.100.07
1
1
-0.33466752
-1
FALSE
19
2143
Allopurinol
All.100.07
1
1
-0.25854402
-1
FALSE
104
2408
Vehicle
Veh.000.03
1
1
-1.4422447
-1
FALSE
121
2413
Vehicle
Veh.000.14
1
1
-1.277754
-1
FALSE
128
1940
Vehicle
Veh.000.14
1
1
-1.2135719
-1
FALSE
116
1936
Vehicle
Veh.000.07
1
1
-1.1563262
-1
FALSE
54
1962
Puromycin
Pur.020.03
2
1
-1.0786993
-1
FALSE
63
1971
Puromycin
Pur.020.14
2
1
-1.3810734
-1
FALSE
61
1969
Puromycin
Pur.020.14
3
1
-0.056223804
-1
FALSE

The mis-classified samples by SVM in Charles River Laboratories studies are listed here. Columns are: A_id, animal identification; Treatment, compound; Cpd.Dose.Day, compound.dose.day; H_score, histopathology grade; B_class, designated SVM class label for testing; SVM value, the prediction value from SVM model (>0 indicates positive class and <0 indicates negative class); class_predicted, predicted class label; Prediction, prediction true or false.

Jiang et al. Journal of Translational Medicine 2007 5:47   doi:10.1186/1479-5876-5-47