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 CentroidFold - scored higher in this pairwise comparison

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

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


Overview

Metric CentroidFold NanoFolder
MCC 0.564 > 0.406
Average MCC ± 95% Confidence Intervals 0.532 ± 0.187 > 0.450 ± 0.163
Sensitivity 0.560 > 0.516
Positive Predictive Value 0.579 > 0.333
Total TP 140 > 129
Total TN 20939 > 20794
Total FP 137 < 311
Total FP CONTRA 32 < 98
Total FP INCONS 70 < 160
Total FP COMP 35 < 53
Total FN 110 < 121
P-value 2.61939086268e-08

^top




Performance plots


  1. Comparison of performance of CentroidFold and NanoFolder. 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 CentroidFold and NanoFolder).

  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 CentroidFold and NanoFolder).

  3. Comparison of average Matthews Correlation Coefficients (MCCs) for CentroidFold and NanoFolder. 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 CentroidFold and NanoFolder).

^top





Performance of CentroidFold - scored higher in this pairwise comparison

1. Total counts & total scores for CentroidFold

Total Base Pair Counts
Total TP 140
Total TN 20939
Total FP 137
Total FP CONTRA 32
Total FP INCONS 70
Total FP COMP 35
Total FN 110
Total Scores
MCC 0.564
Average MCC ± 95% Confidence Intervals 0.532 ± 0.187
Sensitivity 0.560
Positive Predictive Value 0.579
Nr of predictions 13

^top



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

RNA Chain Rfam family MCC SENS PPV TP TN FP FP CONTRA FP INCONS FP COMP FN
2LC8_A 0.51 0.39 0.70 7 518 3 1 2 0 11
2M58_A - -0.02 0.00 0.00 0 538 6 0 6 0 12
3J3D_C 0.73 0.79 0.68 15 946 7 3 4 0 4
3J3E_8 0.12 0.13 0.13 2 2726 22 4 10 8 13
3J3F_8 0.36 0.47 0.28 9 4729 37 12 11 14 10
3U4M_B - 0.81 0.77 0.85 17 1256 5 0 3 2 5
3W3S_B 0.95 0.91 1.00 30 1959 1 0 0 1 3
4A1C_2 0.22 0.25 0.19 5 4490 29 9 12 8 15
4AOB_A 0.50 0.48 0.54 14 1411 13 3 9 1 15
4ATO_G - 0.61 0.57 0.67 4 214 2 0 2 0 3
4ENC_A 0.85 0.73 1.00 11 485 1 0 0 1 4
4JF2_A 0.89 0.79 1.00 19 1063 0 0 0 0 5
4JRC_A - 0.38 0.41 0.39 7 604 11 0 11 0 10

^top



Performance of NanoFolder - scored lower in this pairwise comparison

1. Total counts & total scores for NanoFolder

Total Base Pair Counts
Total TP 129
Total TN 20794
Total FP 311
Total FP CONTRA 98
Total FP INCONS 160
Total FP COMP 53
Total FN 121
Total Scores
MCC 0.406
Average MCC ± 95% Confidence Intervals 0.450 ± 0.163
Sensitivity 0.516
Positive Predictive Value 0.333
Nr of predictions 13

^top



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

RNA Chain Rfam family MCC SENS PPV TP TN FP FP CONTRA FP INCONS FP COMP FN
2LC8_A 0.54 0.61 0.50 11 506 11 1 10 0 7
2M58_A - 0.45 0.58 0.37 7 525 13 6 6 1 5
3J3D_C 0.76 0.95 0.62 18 939 11 9 2 0 1
3J3E_8 -0.01 0.00 0.00 0 2707 47 11 24 12 15
3J3F_8 0.30 0.47 0.19 9 4714 57 21 17 19 10
3U4M_B - 0.77 0.91 0.67 20 1246 12 6 4 2 2
3W3S_B 0.18 0.21 0.18 7 1949 34 2 31 1 26
4A1C_2 -0.01 0.00 0.00 0 4469 61 18 29 14 20
4AOB_A 0.39 0.41 0.39 12 1406 20 4 15 1 17
4ATO_G - 0.72 1.00 0.54 7 207 6 6 0 0 0
4ENC_A 0.66 0.73 0.61 11 478 9 2 5 2 4
4JF2_A 0.55 0.67 0.47 16 1048 19 9 9 1 8
4JRC_A - 0.56 0.65 0.50 11 600 11 3 8 0 6

^top


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