CompaRNA - on-line benchmarks of RNA structure prediction methods
Home

Methods
Datasets
Rankings
RNA 2D Atlas

Help
FAQ

Contact us
RSS feeds
Twitter

Table of contents:

  1. Overview

  2. Performance Plots

  3. Performance of RNAalifold(20) - scored higher in this pairwise comparison

  4. Performance of Fold - scored lower in this pairwise comparison

  5. Compile and download dataset for RNAalifold(20) & Fold [.zip] - may take several seconds...


Overview

Metric RNAalifold(20) Fold
MCC 0.646 > 0.581
Average MCC ± 95% Confidence Intervals 0.642 ± 0.107 > 0.567 ± 0.118
Sensitivity 0.494 < 0.512
Positive Predictive Value 0.847 > 0.662
Total TP 333 < 345
Total TN 154823 > 154695
Total FP 73 < 213
Total FP CONTRA 14 < 24
Total FP INCONS 46 < 152
Total FP COMP 13 < 37
Total FN 341 > 329
P-value 1.82627677697e-08

^top




Performance plots


  1. Comparison of performance of RNAalifold(20) and Fold. 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 RNAalifold(20) and Fold).

  2. 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 RNAalifold(20) and Fold).

  3. Comparison of average Matthews Correlation Coefficients (MCCs) for RNAalifold(20) and Fold. 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 RNAalifold(20) and Fold).

^top





Performance of RNAalifold(20) - scored higher in this pairwise comparison

1. Total counts & total scores for RNAalifold(20)

Total Base Pair Counts
Total TP 333
Total TN 154823
Total FP 73
Total FP CONTRA 14
Total FP INCONS 46
Total FP COMP 13
Total FN 341
Total Scores
MCC 0.646
Average MCC ± 95% Confidence Intervals 0.642 ± 0.107
Sensitivity 0.494
Positive Predictive Value 0.847
Nr of predictions 15

^top



2. Individual counts for RNAalifold(20) [ download as .csv ]

RNA Chain Rfam family MCC SENS PPV TP TN FP FP CONTRA FP INCONS FP COMP FN
2XQD_Y 0.88 0.78 1.00 21 2829 0 0 0 0 6
3AMU_B 0.86 0.74 1.00 20 2983 1 0 0 1 7
3IZ4_A 0.53 0.35 0.82 46 70820 15 4 6 5 86
3IZF_C 0.71 0.59 0.86 32 6866 5 1 4 0 22
3NPB_A 0.68 0.48 0.96 22 6998 3 1 0 2 24
3O58_3 0.37 0.26 0.53 9 12386 8 2 6 0 26
3O58_2 0.79 0.76 0.83 29 7225 7 3 3 1 9
3PDR_A 0.76 0.61 0.96 44 12834 3 0 2 1 28
3RKF_A 0.68 0.50 0.94 17 2193 1 0 1 0 17
3SD1_A 0.70 0.60 0.83 25 3886 5 1 4 0 17
4A1C_2 0.17 0.15 0.19 5 11755 23 2 19 2 28
4A1C_3 0.74 0.57 0.97 31 7108 1 0 1 0 23
4AOB_A 0.72 0.52 1.00 22 4349 1 0 0 1 20
4ENB_A 0.46 0.21 1.00 4 1271 0 0 0 0 15
4ENC_A 0.56 0.32 1.00 6 1320 0 0 0 0 13

^top



Performance of Fold - scored lower in this pairwise comparison

1. Total counts & total scores for Fold

Total Base Pair Counts
Total TP 345
Total TN 154695
Total FP 213
Total FP CONTRA 24
Total FP INCONS 152
Total FP COMP 37
Total FN 329
Total Scores
MCC 0.581
Average MCC ± 95% Confidence Intervals 0.567 ± 0.118
Sensitivity 0.512
Positive Predictive Value 0.662
Nr of predictions 15

^top



2. Individual counts for Fold [ download as .csv ]

RNA Chain Rfam family MCC SENS PPV TP TN FP FP CONTRA FP INCONS FP COMP FN
2XQD_Y 0.88 0.78 1.00 21 2829 0 0 0 0 6
3AMU_B 0.64 0.59 0.70 16 2980 9 0 7 2 11
3IZ4_A 0.53 0.46 0.61 61 70776 44 5 34 5 71
3IZF_C 0.70 0.59 0.82 32 6864 7 1 6 0 22
3NPB_A 0.70 0.61 0.80 28 6986 10 1 6 3 18
3O58_3 0.29 0.31 0.28 11 12363 41 3 26 12 24
3O58_2 0.71 0.71 0.71 27 7222 12 3 8 1 11
3PDR_A 0.77 0.64 0.94 46 12831 5 0 3 2 26
3RKF_A 0.76 0.59 1.00 20 2191 0 0 0 0 14
3SD1_A 0.57 0.48 0.69 20 3887 9 1 8 0 22
4A1C_2 0.14 0.15 0.14 5 11744 43 5 27 11 28
4A1C_3 0.68 0.57 0.82 31 7102 7 1 6 0 23
4AOB_A 0.50 0.40 0.63 17 4344 11 2 8 1 25
4ENB_A 0.32 0.26 0.42 5 1263 7 1 6 0 14
4ENC_A 0.31 0.26 0.38 5 1313 8 1 7 0 14

^top


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