CompaRNA - on-line benchmarks of RNA structure prediction methods

FAQ

Metric	PETfold_pre2.0(seed)		RNAalifold(seed)
MCC	0.680	>	0.637
Average MCC ± 95% Confidence Intervals	0.625 ± 0.077	>	0.515 ± 0.142
Sensitivity	0.570	>	0.477
Positive Predictive Value	0.811	<	0.849
Total TP	607	>	508
Total TN	1190681	<	1190831
Total FP	168	>	98
Total FP CONTRA	17	>	6
Total FP INCONS	124	>	84
Total FP COMP	27	>	8
Total FN	457	<	556
P-value	1.91441904741e-08

Comparison of performance of PETfold_pre2.0(seed) and RNAalifold(seed). 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 PETfold_pre2.0(seed) and RNAalifold(seed)).
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 PETfold_pre2.0(seed) and RNAalifold(seed)).
Comparison of average Matthews Correlation Coefficients (MCCs) for PETfold_pre2.0(seed) and RNAalifold(seed). 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 PETfold_pre2.0(seed) and RNAalifold(seed)).

Total Scores
MCC	0.680
Average MCC ± 95% Confidence Intervals	0.625 ± 0.077
Sensitivity	0.570
Positive Predictive Value	0.811
Nr of predictions	14

RNA Chain	Rfam family	MCC	SENS	PPV	TP	TN	FP	FP CONTRA	FP INCONS	FP COMP	FN
2LC8_A	GP_knot1 links to Rfam database...	0.40	0.30	0.55	6	1529	5	0	5	0	14
3J20_2	SSU_rRNA_archaea links to Rfam database...	0.71	0.61	0.81	388	1116287	97	10	80	7	245
3J3E_8	5_8S_rRNA links to Rfam database...	0.43	0.30	0.63	10	7487	8	0	6	2	23
3J3F_8	5_8S_rRNA links to Rfam database...	0.63	0.53	0.76	19	12221	10	0	6	4	17
3W1K_J	tRNA-Sec links to Rfam database...	0.80	0.68	0.93	26	4158	3	1	1	1	12
3W3S_B	tRNA-Sec links to Rfam database...	0.78	0.65	0.93	26	4725	4	0	2	2	14
3ZEX_C	5_8S_rRNA links to Rfam database...	0.53	0.37	0.76	19	14171	10	1	5	4	33
4A1C_2	5_8S_rRNA links to Rfam database...	0.57	0.45	0.71	15	11760	12	0	6	6	18
4AOB_A	SAM links to Rfam database...	0.72	0.57	0.92	24	4345	3	0	2	1	18
4ENB_A	crcB links to Rfam database...	0.50	0.42	0.62	8	1262	5	1	4	0	11
4ENC_A	crcB links to Rfam database...	0.50	0.42	0.62	8	1313	5	1	4	0	11
4FRG_B	AdoCbl-variant links to Rfam database...	0.74	0.63	0.87	20	3463	3	1	2	0	12
4FRN_A	AdoCbl-variant links to Rfam database...	0.73	0.61	0.88	22	5126	3	2	1	0	14
4JF2_A	preQ1-II links to Rfam database...	0.72	0.52	1.00	16	2834	0	0	0	0	15

Total Scores
MCC	0.637
Average MCC ± 95% Confidence Intervals	0.515 ± 0.142
Sensitivity	0.477
Positive Predictive Value	0.849
Nr of predictions	14

RNA Chain	Rfam family	MCC	SENS	PPV	TP	TN	FP	FP CONTRA	FP INCONS	FP COMP	FN
2LC8_A	GP_knot1 links to Rfam database...	-0.01	0.00	0.00	0	1528	12	0	12	0	20
3J20_2	SSU_rRNA_archaea links to Rfam database...	0.69	0.58	0.83	367	1116322	80	5	71	4	266
3J3E_8	5_8S_rRNA links to Rfam database...	0.00	0.00	0.00	0	7503	0	0	0	0	33
3J3F_8	5_8S_rRNA links to Rfam database...	0.50	0.25	1.00	9	12237	0	0	0	0	27
3W1K_J	tRNA-Sec links to Rfam database...	0.71	0.50	1.00	19	4167	0	0	0	0	19
3W3S_B	tRNA-Sec links to Rfam database...	0.67	0.45	1.00	18	4735	1	0	0	1	22
3ZEX_C	5_8S_rRNA links to Rfam database...	0.42	0.17	1.00	9	14187	0	0	0	0	43
4A1C_2	5_8S_rRNA links to Rfam database...	0.39	0.15	1.00	5	11776	2	0	0	2	28
4AOB_A	SAM links to Rfam database...	0.72	0.52	1.00	22	4349	1	0	0	1	20
4ENB_A	crcB links to Rfam database...	0.56	0.32	1.00	6	1269	0	0	0	0	13
4ENC_A	crcB links to Rfam database...	0.56	0.32	1.00	6	1320	0	0	0	0	13
4FRG_B	AdoCbl-variant links to Rfam database...	0.73	0.53	1.00	17	3469	0	0	0	0	15
4FRN_A	AdoCbl-variant links to Rfam database...	0.69	0.53	0.90	19	5130	2	1	1	0	17
4JF2_A	preQ1-II links to Rfam database...	0.59	0.35	1.00	11	2839	0	0	0	0	20

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