Table 4.

Averaged AUCs from our dataset for 100 hold-out validation runs of our machine learning algorithms.

	Validation set				Test set
Type of feature set	RF	NB	XGB	LR	RF	NB	XGB	LR
All features	0.9009	0.8713	0.8917	0.8641	0.9018	0.8713	0.8921	0.8642
Base + DDIs-N + SNPs + DDIs-D	0.8964	0.8575	0.8834	0.8501	0.8973	0.8575	0.8835	0.8501
Base + SE-AH + SNPs + DDIs-D	0.8961	0.8710	0.8896	0.8655	0.8970	0.8710	0.8896	0.8656
Base + SE-AH + DDIs-N + DDIs-D	0.8947	0.8645	0.8859	0.8550	0.8959	0.8644	0.8863	0.8550
Base + SNPs + DDIs-N + SE-AH	0.8940	0.8645	0.8868	0.8563	0.8951	0.8644	0.8870	0.8563
Base + SNPs + DDIs-N	0.8905	0.8572	0.8809	0.8519	0.8913	0.8572	0.8809	0.8519
Base + DDIs-D + DDIs-N	0.8901	0.8505	0.8773	0.8400	0.8911	0.8505	0.8773	0.8401
Base + SNPs + DDIs-N	0.8886	0.8496	0.8775	0.8415	0.8897	0.8496	0.8776	0.8414
Base + DDIs-D + SE-AH	0.8879	0.8641	0.8830	0.8563	0.8890	0.8640	0.8830	0.8563
Base + SE-AH + SNPs	0.8874	0.8641	0.8840	0.8574	0.8885	0.8690	0.8840	0.8575
Base + SE-AH + DDIs-N	0.8849	0.8540	0.8783	0.8422	0.8863	0.8539	0.8785	0.8421
Base + DDIs-D	0.8812	0.8502	0.8736	0.8432	0.8820	0.8502	0.8736	0.8431
Base + SNPs	0.8798	0.8493	0.8744	0.8440	0.8810	0.8492	0.8744	0.8440
Base + DDIs-N	0.8788	0.8389	0.8681	0.8281	0.8802	0.8389	0.8683	0.8279
Base + SE-AH	0.8761	0.8535	0.8740	0.8430	0.8774	0.8533	0.8741	0.8429
Base	0.8659	0.8385	0.8634	0.8313	0.8673	0.8385	0.8636	0.8310