Skip to main content

Table 4 A table showing the top 10 mutation positions that have the most significant impact on the algorithms across all drags

From: Generalizability of machine learning in predicting antimicrobial resistance in E. coli: a multi-country case study in Africa

 

Logistic Regression

Random Forest

Gradient Boosting

XGBoost

LightGBM

CatBoost

Feed-Forward NN

SVM

 

(‘Feature’, ‘Score’)

(‘Feature’, ‘Score’)

(‘Feature’, ‘Score’)

(‘Feature’, ‘Score’)

(‘Feature’, ‘Score’)

(‘Feature’, ‘Score’)

(‘Feature’, ‘Score’)

(‘Feature’, ‘Score’)

1

(‘3,589,009’, 0.47)

(‘1,008,149’, 0.002)

(‘2,189,699’, 0.05)

(‘4,374,394’, 0.04)

(‘4,040,529’, 27)

(‘4,428,325’, 2.39)

(‘4,614,839’, 0.03)

(‘1,008,149’, 151.24)

2

(‘4,040,529’, 0.39)

(‘3,284,134’, 0.004)

(‘3,230,190’, 0.05)

(‘1,008,149’, 0.03)

(‘4,437,351’, 27)

(‘3,230,190’, 1.5)

(‘4,428,325’, 0.02)

(‘3,009,381’, 131.13)

3

(‘1,473,047’, 0.34)

(‘4,240,504’, 0.002)

(‘1,008,149’, 0.05)

(‘3,284,134’, 0.02)

(‘2,160,560’, 26)

(‘2,189,699’, 1.38)

(‘4,615,406’, 0.02)

(‘4,172,848’, 119.69)

4

(‘3,088,751’, 0.32)

(‘1,263,246’, 0.003)

(‘3,284,134’, 0.04)

(‘3,615,876’, 0.02)

(‘3,680,451’, 25)

(‘1,008,149’, 1.26)

(‘4,437,351’, 0.02)

(‘4,445,374’, 108.66)

5

(‘3,117,412’, 0.31)

(‘4,453,756’, 0.002)

(‘1,263,246’, 0.03)

(‘2,189,699’, 0.02)

(‘3,118,273’, 21)

(‘3,284,134’, 1.08)

(‘4,458,693’, 0.02)

(‘3,418,178’, 107.1)

6

(‘3,118,273’, 0.31)

(‘2,824,686’, 0.002)

(‘4,240,504’, 0.03)

(‘1,239,691’, 0.01)

(‘4,428,325’, 21)

(‘4,615,406’, 1.08)

(‘4,452,347’, 0.02)

(‘2,791,632’, 106.19)

7

(‘1,324,783’, 0.3)

(‘4,374,394’, 0.003)

(‘4,374,394’, 0.03)

(‘3,230,190’, 0.01)

(‘4,481,654’, 21)

(‘3,584,376’, 0.99)

(‘3,200,153’, 0.02)

(‘1,264,526’, 104.35)

8

(‘1,720,453’, 0.28)

(‘4,286,100’, 0.002)

(‘3,373,941’, 0.03)

(‘2,792,739’, 0.01)

(‘2,480,806’, 20)

(‘4,445,374’, 0.97)

(‘4,443,644’, 0.02)

(‘4,172,776’, 103.49)

9

(‘4,446,973’, 0.28)

(‘3,614,716’, 0.002)

(‘3,615,876’, 0.02)

(‘3,021,822’, 0.01)

(‘3,589,009’, 18)

(‘3,681,794’, 0.96)

(‘4,434,509’, 0.02)

(‘4,218,643’, 102.63)

10

(‘2,778,469’, 0.28)

(‘1,466,504’, 0.002)

(‘3,681,794’, 0.02)

(‘3,373,941’, 0.01)

(‘2,035,257’, 16)

(‘4,269,331’, 0.96)

(‘3,584,376’, 0.02)

(‘3,300,216’, 100.93)