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 Cylofold - scored higher in this pairwise comparison

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

  5. Compile and download dataset for Cylofold & RNAwolf [.zip] - may take several seconds...


Overview

Metric Cylofold RNAwolf
MCC 0.537 > 0.451
Average MCC ± 95% Confidence Intervals 0.516 ± 0.140 > 0.410 ± 0.137
Sensitivity 0.454 > 0.411
Positive Predictive Value 0.643 > 0.506
Total TP 189 > 171
Total TN 41992 > 41948
Total FP 110 < 179
Total FP CONTRA 9 < 16
Total FP INCONS 96 < 151
Total FP COMP 5 < 12
Total FN 227 < 245
P-value 1.89649746203e-08

^top




Performance plots


  1. Comparison of performance of Cylofold and RNAwolf. 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 Cylofold and RNAwolf).

  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 Cylofold and RNAwolf).

  3. Comparison of average Matthews Correlation Coefficients (MCCs) for Cylofold and RNAwolf. 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 Cylofold and RNAwolf).

^top





Performance of Cylofold - scored higher in this pairwise comparison

1. Total counts & total scores for Cylofold

Total Base Pair Counts
Total TP 189
Total TN 41992
Total FP 110
Total FP CONTRA 9
Total FP INCONS 96
Total FP COMP 5
Total FN 227
Total Scores
MCC 0.537
Average MCC ± 95% Confidence Intervals 0.516 ± 0.140
Sensitivity 0.454
Positive Predictive Value 0.643
Nr of predictions 15

^top



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

RNA Chain Rfam family MCC SENS PPV TP TN FP FP CONTRA FP INCONS FP COMP FN
2LC8_A 0.61 0.55 0.69 11 1524 5 0 5 0 9
2LKR_A - 0.50 0.44 0.59 17 6076 12 2 10 0 22
3AMU_B 0.67 0.59 0.76 16 2982 7 0 5 2 11
3J0L_2 - 0.39 0.36 0.43 12 6188 18 2 14 2 21
3J0L_g - -0.01 0.00 0.00 0 461 4 1 3 0 4
3J0L_a - 0.22 0.19 0.27 3 1117 8 1 7 0 13
3J0L_7 - 0.30 0.29 0.33 5 1210 10 0 10 0 12
3J16_L 0.75 0.57 1.00 17 2758 0 0 0 0 13
3RKF_A 0.76 0.59 1.00 20 2191 0 0 0 0 14
3SD1_A 0.62 0.50 0.78 21 3889 6 0 6 0 21
3U4M_B - 0.38 0.32 0.46 12 3134 14 0 14 0 25
3UZL_B 0.45 0.38 0.54 14 3544 12 1 11 0 23
4AOB_A 0.35 0.26 0.48 11 4348 13 1 11 1 31
4ENB_A 0.89 0.79 1.00 15 1260 0 0 0 0 4
4ENC_A 0.86 0.79 0.94 15 1310 1 1 0 0 4

^top



Performance of RNAwolf - scored lower in this pairwise comparison

1. Total counts & total scores for RNAwolf

Total Base Pair Counts
Total TP 171
Total TN 41948
Total FP 179
Total FP CONTRA 16
Total FP INCONS 151
Total FP COMP 12
Total FN 245
Total Scores
MCC 0.451
Average MCC ± 95% Confidence Intervals 0.410 ± 0.137
Sensitivity 0.411
Positive Predictive Value 0.506
Nr of predictions 15

^top



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

RNA Chain Rfam family MCC SENS PPV TP TN FP FP CONTRA FP INCONS FP COMP FN
2LC8_A 0.52 0.50 0.56 10 1522 8 0 8 0 10
2LKR_A - 0.81 0.74 0.88 29 6072 8 0 4 4 10
3AMU_B 0.68 0.63 0.74 17 2980 9 0 6 3 10
3J0L_2 - 0.11 0.12 0.11 4 6181 34 5 26 3 29
3J0L_g - 0.16 0.25 0.11 1 456 8 6 2 0 3
3J0L_a - 0.18 0.19 0.20 3 1113 12 1 11 0 13
3J0L_7 - -0.01 0.00 0.00 0 1214 11 0 11 0 17
3J16_L 0.45 0.40 0.52 12 2752 11 0 11 0 18
3RKF_A 0.72 0.62 0.84 21 2186 4 0 4 0 13
3SD1_A 0.46 0.43 0.50 18 3880 18 0 18 0 24
3U4M_B - 0.44 0.38 0.52 14 3133 13 0 13 0 23
3UZL_B 0.72 0.59 0.88 22 3545 4 1 2 1 15
4AOB_A 0.23 0.19 0.30 8 4344 20 1 18 1 34
4ENB_A 0.35 0.32 0.40 6 1260 9 1 8 0 13
4ENC_A 0.34 0.32 0.38 6 1310 10 1 9 0 13

^top


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