MolData, a molecular benchmark for disease and target based machine learning

Table 3 Target-based information for the MolData Benchmark

Target	AID count	Unique target count	Active data points	Total data points	% Active datapoints	Unique active molecules	Total unique molecules	% Unique active molecules
All Targets	383	296	862,370	103,440,515	0.83	261,715	675,161	38.76
Membrane receptor	85	44	146,956	25,922,533	0.56	91,489	458,818	19.94
Enzyme (other)	54	51	83,657	16,210,090	0.51	57,808	632,142	9.14
Nuclear receptor	53	25	74,776	6,083,509	1.22	42,838	442,487	9.68
Hydrolase	36	32	113,185	10,830,324	1.05	66,195	526,391	12.57
Protease	29	26	37,943	7,965,313	0.47	30,619	606,793	5.05
Transcription factor	27	18	53,416	4,775,685	1.11	40,067	503,249	7.96
Kinase	24	23	38,257	7,369,690	0.52	31,327	377,519	8.29
Epigenetic regulator	23	20	76,793	6,840,095	1.12	51,776	523,904	9.88
Ion channel	22	14	37,402	6,745,762	0.55	28,853	511,873	5.63
Transferase	18	17	43,955	6,279,651	0.7	30,432	519,646	5.85
Oxidoreductase	10	8	33,956	2,953,760	1.15	30,054	432,578	6.94
Transporter	9	8	15,390	2,538,579	0.60	15,046	369,621	4.07
NTPase	6	5	114,465	1,981,575	5.78	76,334	439,967	17.34
Phosphatase	5	5	8090	1,693,773	0.48	6913	368,329	1.87

ISSN: 1758-2946