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 Fold - scored higher in this pairwise comparison

  4. Performance of Multilign(20) - scored lower in this pairwise comparison

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


Overview

Metric Fold Multilign(20)
MCC 0.703 > 0.652
Average MCC ± 95% Confidence Intervals 0.677 ± 0.157 > 0.638 ± 0.177
Sensitivity 0.712 > 0.624
Positive Predictive Value 0.703 > 0.692
Total TP 267 > 234
Total TN 24611 < 24653
Total FP 150 > 131
Total FP CONTRA 30 > 23
Total FP INCONS 83 > 81
Total FP COMP 37 > 27
Total FN 108 < 141
P-value 5.03356661821e-08

^top




Performance plots


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

  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 Fold and Multilign(20)).

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

^top





Performance of Fold - scored higher in this pairwise comparison

1. Total counts & total scores for Fold

Total Base Pair Counts
Total TP 267
Total TN 24611
Total FP 150
Total FP CONTRA 30
Total FP INCONS 83
Total FP COMP 37
Total FN 108
Total Scores
MCC 0.703
Average MCC ± 95% Confidence Intervals 0.677 ± 0.157
Sensitivity 0.712
Positive Predictive Value 0.703
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
2L94_A 1.00 1.00 1.00 18 339 1 0 0 1 0
2XXA_G 1.00 1.00 1.00 35 2010 1 0 0 1 0
3AMU_B 0.73 0.79 0.68 15 1135 10 0 7 3 4
3J20_0 0.54 0.57 0.52 12 1196 12 3 8 1 9
3J20_1 1.00 1.00 1.00 20 1092 1 0 0 1 0
3J2L_3 0.80 0.82 0.78 28 2984 12 1 7 4 6
3RKF_A 0.91 0.83 1.00 20 846 0 0 0 0 4
3SD1_A 0.68 0.69 0.69 20 1504 9 4 5 0 9
3ZEX_D 0.90 0.86 0.94 30 2764 8 0 2 6 5
4A1C_3 0.86 0.84 0.89 31 2728 7 0 4 3 6
4A1C_2 0.19 0.25 0.15 5 4482 43 11 18 14 15
4AOB_A 0.60 0.59 0.63 17 1410 11 4 6 1 12
4ENB_A 0.37 0.33 0.45 5 461 7 0 6 1 10
4ENC_A 0.36 0.33 0.42 5 484 8 0 7 1 10
4FRG_B 0.22 0.25 0.23 6 1176 20 7 13 0 18

^top



Performance of Multilign(20) - scored lower in this pairwise comparison

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

Total Base Pair Counts
Total TP 234
Total TN 24653
Total FP 131
Total FP CONTRA 23
Total FP INCONS 81
Total FP COMP 27
Total FN 141
Total Scores
MCC 0.652
Average MCC ± 95% Confidence Intervals 0.638 ± 0.177
Sensitivity 0.624
Positive Predictive Value 0.692
Nr of predictions 15

^top



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

RNA Chain Rfam family MCC SENS PPV TP TN FP FP CONTRA FP INCONS FP COMP FN
2L94_A 1.00 1.00 1.00 18 339 1 0 0 1 0
2XXA_G -0.01 0.00 0.00 0 2033 12 2 10 0 35
3AMU_B 0.71 0.74 0.70 14 1137 9 0 6 3 5
3J20_0 0.68 0.71 0.65 15 1196 9 2 6 1 6
3J20_1 1.00 1.00 1.00 20 1092 1 0 0 1 0
3J2L_3 0.80 0.79 0.82 27 2987 10 0 6 4 7
3RKF_A 0.91 0.88 0.95 21 844 1 0 1 0 3
3SD1_A 0.68 0.69 0.69 20 1504 9 4 5 0 9
3ZEX_D 0.88 0.83 0.94 29 2765 8 0 2 6 6
4A1C_3 0.88 0.86 0.89 32 2727 6 0 4 2 5
4A1C_2 0.20 0.25 0.16 5 4485 33 9 17 7 15
4AOB_A 0.59 0.55 0.64 16 1412 10 3 6 1 13
4ENB_A 0.39 0.33 0.50 5 462 6 0 5 1 10
4ENC_A 0.73 0.60 0.90 9 486 1 1 0 0 6
4FRG_B 0.13 0.13 0.17 3 1184 15 2 13 0 21

^top


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