CompaRNA - on-line benchmarks of RNA structure prediction methods

FAQ

Metric	CentroidAlifold(20)		PknotsRG
MCC	0.803	>	0.664
Average MCC ± 95% Confidence Intervals	0.799 ± 0.092	>	0.684 ± 0.172
Sensitivity	0.694	>	0.685
Positive Predictive Value	0.932	>	0.649
Total TP	315	>	311
Total TN	51925	>	51784
Total FP	39	<	206
Total FP CONTRA	16	<	62
Total FP INCONS	7	<	106
Total FP COMP	16	<	38
Total FN	139	<	143
P-value	2.86595104665e-08

Comparison of performance of CentroidAlifold(20) and PknotsRG. Positive Predictive Value (PPV) is plotted against sensitivity. Each dot represents a single test of each method. See tables below for raw data (individual counts for CentroidAlifold(20) and PknotsRG).
Average Matthews Correlation Coefficients (MCC) with 95% confidence intervals (CIs) were plotted for different RNA families, for which at least 3 members were present in the benchmarking dataset. 'n' denotes the number of MCCs used to calculate the average and CI. See tables below for raw data (individual counts for CentroidAlifold(20) and PknotsRG).
Comparison of average Matthews Correlation Coefficients (MCCs) for CentroidAlifold(20) and PknotsRG. The whiskers correspond to 95% confidence intervals (CIs). 'n' denotes the number of MCCs used to calculate average MCCs and CIs. See tables below for raw data (individual counts for CentroidAlifold(20) and PknotsRG).

Total Scores
MCC	0.803
Average MCC ± 95% Confidence Intervals	0.799 ± 0.092
Sensitivity	0.694
Positive Predictive Value	0.932
Nr of predictions	15

RNA Chain	Rfam family	MCC	SENS	PPV	TP	TN	FP	FP CONTRA	FP INCONS	FP COMP	FN
3A2K_C	tRNA links to Rfam database...	0.98	0.95	1.00	21	1087	0	0	0	0	1
3GX2_A	SAM links to Rfam database...	0.92	0.86	1.00	24	1425	1	0	0	1	4
3IVN_B	Purine links to Rfam database...	0.86	0.83	0.90	19	882	2	2	0	0	4
3IZ4_A	tmRNA links to Rfam database...	0.67	0.52	0.88	49	25480	9	7	0	2	46
3JYV_7	tRNA links to Rfam database...	0.97	0.95	1.00	19	1092	2	0	0	2	1
3LA5_A	Purine links to Rfam database...	0.87	0.80	0.95	20	933	1	1	0	0	5
3NPB_A	SAM links to Rfam database...	0.77	0.65	0.92	24	2252	6	1	1	4	13
3O58_3	5_8S_rRNA links to Rfam database...	0.64	0.45	0.91	10	4753	2	1	0	1	12
3PDR_A	ykoK links to Rfam database...	0.92	0.90	0.94	45	4792	5	1	2	2	5
3RKF_A	Purine links to Rfam database...	0.86	0.75	1.00	18	848	0	0	0	0	6
3SD1_A	THF links to Rfam database...	0.85	0.76	0.96	22	1510	1	0	1	0	7
4A1C_2	5_8S_rRNA links to Rfam database...	0.33	0.25	0.45	5	4505	8	3	3	2	15
4AOB_A	SAM links to Rfam database...	0.89	0.79	1.00	23	1414	2	0	0	2	6
4ENB_A	crcB links to Rfam database...	0.77	0.60	1.00	9	463	0	0	0	0	6
4ENC_A	crcB links to Rfam database...	0.68	0.47	1.00	7	489	0	0	0	0	8

Total Scores
MCC	0.664
Average MCC ± 95% Confidence Intervals	0.684 ± 0.172
Sensitivity	0.685
Positive Predictive Value	0.649
Nr of predictions	15

RNA Chain	Rfam family	MCC	SENS	PPV	TP	TN	FP	FP CONTRA	FP INCONS	FP COMP	FN
3A2K_C	tRNA links to Rfam database...	0.47	0.50	0.46	11	1084	13	3	10	0	11
3GX2_A	SAM links to Rfam database...	0.80	0.79	0.81	22	1422	6	2	3	1	6
3IVN_B	Purine links to Rfam database...	0.91	0.83	1.00	19	884	0	0	0	0	4
3IZ4_A	tmRNA links to Rfam database...	0.56	0.60	0.52	57	25426	57	25	28	4	38
3JYV_7	tRNA links to Rfam database...	-0.02	0.00	0.00	0	1089	22	4	18	0	20
3LA5_A	Purine links to Rfam database...	0.91	0.84	1.00	21	933	0	0	0	0	4
3NPB_A	SAM links to Rfam database...	0.93	0.86	1.00	32	2246	4	0	0	4	5
3O58_3	5_8S_rRNA links to Rfam database...	0.41	0.50	0.34	11	4732	35	6	15	14	11
3PDR_A	ykoK links to Rfam database...	0.90	0.90	0.90	45	4790	7	2	3	2	5
3RKF_A	Purine links to Rfam database...	0.91	0.83	1.00	20	846	0	0	0	0	4
3SD1_A	THF links to Rfam database...	0.73	0.72	0.75	21	1505	7	4	3	0	8
4A1C_2	5_8S_rRNA links to Rfam database...	0.18	0.25	0.14	5	4480	43	11	20	12	15
4AOB_A	SAM links to Rfam database...	0.60	0.59	0.63	17	1410	11	4	6	1	12
4ENB_A	crcB links to Rfam database...	1.00	1.00	1.00	15	457	0	0	0	0	0
4ENC_A	crcB links to Rfam database...	0.97	1.00	0.94	15	480	1	1	0	0	0

Matthews Correlation Coeffient, Sensitivity and Positive Predictive Value have been calculated based on the paper by Gardener & Giegerich, 2004.