Table 1.

Table representing the results of the benchmark on the EMT test dataset (A) and on the GM12878 test dataset (B) using a threshold for the PSI values and IR ratios of 0.10

Method	Excl.	TP	TN	FP	FN	TPR	TNR	PPV	Acc.	FDR
A. EMT test results
IRFinder	25989	695	95198	428	138	0.83	0.99	0.62	0.98	0.38
IRFinder-S	25989	673	35515	111	160	0.81	1.00	0.86	0.99	0.14
iREAD	28221	18	33994	86	129	0.12	1.00	0.17	0.99	0.83
Whippet	59822	443	1978	87	118	0.79	0.96	0.84	0.92	0.16
MAJIQ	30179	388	29572	1951	358	0.52	0.94	0.17	0.93	0.83
B. GM12878 test results
IRFinder	30943	1228	50720	729	185	0.87	0.99	0.63	0.98	0.37
IRFinder-S	30943	1077	51123	326	336	0.76	0.99	0.77	0.99	0.23
iREAD	37905	71	45501	179	149	0.32	1.00	0.28	0.99	0.72
Whippet	80125	772	2459	347	102	0.88	0.88	0.69	0.88	0.31
MAJIQ	50932	826	30626	917	504	0.62	0.97	0.47	0.96	0.53

Excl intron excluded, TP true positive, TN true negative, FP false positive, FN false negative, TPR true positive rate (sensitivity), TNR true negative rate (specificity), PPV positive predicted value (precision), Acc. accuracy, FDR false discovery rate