FP-ADMET: a compendium of fingerprint-based ADMET prediction models

Venkatraman, Vishwesh

doi:10.1186/s13321-021-00557-5

Journal of Cheminformatics

Table 4 Performance metrics for the best performing fingerprint-based classification models

From: FP-ADMET: a compendium of fingerprint-based ADMET prediction models

Endpoint	FP	Calibration		Validation
Endpoint	FP	BACC	AUC	BACC	AUC
Blood brain barrier	PUBCHEM	0.82	0.90	0.81	0.92
Oral bioavailability	PUBCHEM	0.71	0.77	0.71	0.78
Anticommensal effect	PUBCHEM	0.76	0.82	0.74	0.81
CYP450 (1A2)	PUBCHEM	0.85	0.93	0.85	0.93
CYP450 (2C19)	ECFP4	0.81	0.88	0.81	0.89
CYP450 (2C9)	PUBCHEM	0.78	0.88	0.79	0.89
CYP450 (2D6)	FCFP4	0.73	0.86	0.73	0.87
CYP450 (3A4)	FCFP6	0.80	0.89	0.80	0.90
CYP450 (2C8)	PUBCHEM	0.79	0.89	0.77	0.90
HIA	MACCS	0.84	0.89	0.83	0.89
BCRP inhibition	FCFP4	0.89	0.95	0.90	0.96
Metabolic intrinsic clearance	FCFP4	0.74	0.82	0.74	0.84
Human liver microsomal stability	AT2D	0.77	0.83	0.77	0.84
PGP inhibitor	PUBCHEM	0.84	0.91	0.85	0.92
PGP substrate	ASP	0.80	0.87	0.80	0.88
DMSO solubility	ECFP2	0.72	0.78	0.73	0.80
Phosphate buffer solubility	PUBCHEM	0.79	0.87	0.79	0.87
Skin sensitization (LLNA)	PUBCHEM	0.69	0.76	0.67	0.74
Skin sensitization (KeratinSens)	LSTAR	0.64	0.65	0.57	0.60
Skin sensitization (HRIPT)	ECFP0	0.70	0.74	0.67	0.72
Skin sensitization (hCLAT)	MACCS	0.65	0.70	0.61	0.68
Skin sensitization (DPRA)	FCFP4	0.68	0.72	0.68	0.72
Rat acute oral toxicity (\(\text {LD}_{{50}}\))	PUBCHEM	0.69	0.78	0.68	0.81
AMES mutagenecity	PUBCHEM	0.79	0.86	0.79	0.87
Cytotoxicity (HepG2)	AT2D	0.78	0.85	0.78	0.85
Cytotoxicity (CRL-7250 cell line)	AT2D	0.79	0.87	0.78	0.86
Cytotoxicity (HACAT cell line)	AT2D	0.77	0.85	0.77	0.85
Cytotoxicity (HEK cell line)	PUBCHEM	0.77	0.87	0.76	0.86
Cytotoxicity (NIK cell line)	PUBCHEM	0.78	0.87	0.78	0.87
DILI	PUBCHEM	0.78	0.86	0.79	0.88
Hemolytic toxicity (saponins)	FCFP6	0.84	0.88	0.85	0.90
hERG cardiotoxicity	FCFP6	0.79	0.86	0.80	0.88
hERG liability	PUBCHEM	0.76	0.87	0.76	0.88
Mitochondrial toxicity	PUBCHEM	0.79	0.90	0.77	0.90
Urinary tract toxicity	FCFP4	0.71	0.77	0.70	0.73
Phototoxicity in vitro	KR	0.70	0.76	0.69	0.80
Phototoxicity human	PUBCHEM	0.69	0.75	0.67	0.75
Toxic myopathy	DFS	0.68	0.74	0.63	0.74
Myelotoxicity	FCFP4	0.72	0.79	0.71	0.80
phospholipidosis	FCFP2	0.78	0.86	0.77	0.88
Cholestasis	RAD2D	0.67	0.73	0.67	0.74
Rhabdomyolysis	MACCS	0.71	0.80	0.70	0.83
Respiratory toxicity	MACCS	0.82	0.88	0.82	0.89
Ototoxicity	PUBCHEM	0.69	0.74	0.67	0.72
MATE1	DFS	0.64	0.67	0.65	0.65
Hepatic steatosis	MACCS	0.63	0.67	0.59	0.68
Carcinogenecity	PUBCHEM	0.67	0.71	0.68	0.75
OATP1B1 inhibition	ECFP6	0.72	0.80	0.73	0.82
OATP2B1 inhibition	ECFP6	0.67	0.68	0.65	0.70
OATP1B3 inhibition	PUBCHEM	0.74	0.83	0.77	0.87
BSEP inhibition	ECFP4	0.85	0.93	0.88	0.95
OCT2 inhibition	PUBCHEM	0.73	0.81	0.73	0.79
PPB	PUBCHEM	0.82	0.92	0.84	0.92
Elimination half-life Human	ASP	0.75	0.86	0.76	0.88
Elimination half-life Mouse	ECFP2	0.74	0.86	0.72	0.84
Elimination half-life Rat	KR	0.74	0.86	0.74	0.83

The values reported are the balanced accuracies (BACC) and area under the ROC curve (AUC) (average of 3 independent runs) for the calibration/validation sets

Back to article page

ISSN: 1758-2946

Contact us

Submission enquiries: journalsubmissions@springernature.com