Skip to main content

Table 1 Comparison of molecules representation methods using PyCaret autoML framework

From: Integrating QSAR modelling with reinforcement learning for Syk inhibitor discovery

Method

Fingerprint category

LightGBM

Random Forest

SVR

Q2

MSE CV

Q2

MSE CV

Q2

MSE CV

ECFP [58]

Circular

0.71

0.37

0.70

0.38

0.72

0.34

Mol2vec [59]

Substructure

0.64

0.46

0.60

0.51

0.47

0.66

MACCS [60]

0.63

0.47

0.61

0.50

0.60

0.49

PubChem fingerprints [61]

0.66

0.44

0.65

0.45

0.62

0.47

MAP4 [62]

String

0.64

0.46

0.61

0.50

0.69

0.38

  1. The bold values highlight the best results within each column, corresponding to the specific model and evaluation metric