Super.FELT: supervised feature extraction learning using triplet loss for drug response prediction with multi-omics data

Table 5 The average AUC scores for 243 drugs in GDSC for the cross validation test

	Super.FELT	MOLIF	AE	ANNF	MOLI	Super.FELT E	Super.FELT M&C	Auto BorutaRF	SVM
Validation AUC with set 1	0.727	0.711	0.712	0.694	0.715	0.723	0.602*
Validation AUC with set 2	0.727	0.697	0.715*	0.699	0.71	0.726	0.598
Validation AUC with set 3	0.728	0.698	0.69	0.697	0.704	0.726	0.597
Validation AUC with set 4	0.73*	0.7137	0.707	0.696	0.694	0.73	0.592	0.747	0.702
Validation AUC with set 5	0.72	0.707	0.672	0.698	0.705	0.727	0.583
Validation AUC with set 6	0.726	0.699	0.708	0.693	0.713	0.724	0.595
Validation AUC with set 7	0.727	0.703	0.712	0.694	0.721*	0.715	0.597
Validation AUC with set 8	0.727	0.7138*	0.684	0.703*	0.719	0.732*	0.593
Test AUC	0.729	0.711	0.719	0.706	0.72	0.728	0.593	0.698	0.7

ISSN: 1471-2105