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 TurboFold(seed) - scored higher in this pairwise comparison

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

  5. Compile and download dataset for TurboFold(seed) & NanoFolder [.zip] - may take several seconds...


Overview

Metric TurboFold(seed) NanoFolder
MCC 0.834 > 0.297
Average MCC ± 95% Confidence Intervals 0.791 ± 0.165 > 0.382 ± 0.206
Sensitivity 0.797 > 0.355
Positive Predictive Value 0.874 > 0.254
Total TP 299 > 133
Total TN 96078 > 95897
Total FP 83 < 421
Total FP CONTRA 5 < 88
Total FP INCONS 38 < 302
Total FP COMP 40 > 31
Total FN 76 < 242
P-value 1.39551335374e-08

^top




Performance plots


  1. Comparison of performance of TurboFold(seed) 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 TurboFold(seed) 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 TurboFold(seed) and NanoFolder).

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

^top





Performance of TurboFold(seed) - scored higher in this pairwise comparison

1. Total counts & total scores for TurboFold(seed)

Total Base Pair Counts
Total TP 299
Total TN 96078
Total FP 83
Total FP CONTRA 5
Total FP INCONS 38
Total FP COMP 40
Total FN 76
Total Scores
MCC 0.834
Average MCC ± 95% Confidence Intervals 0.791 ± 0.165
Sensitivity 0.797
Positive Predictive Value 0.874
Nr of predictions 13

^top



2. Individual counts for TurboFold(seed) [ download as .csv ]

RNA Chain Rfam family MCC SENS PPV TP TN FP FP CONTRA FP INCONS FP COMP FN
CRW_01535 0.94 0.89 1.00 33 7107 3 0 0 3 4
RFA_00416 1.00 1.00 1.00 15 1470 2 0 0 2 0
RFA_00664 -0.01 0.00 0.00 0 980 11 0 10 1 14
RFA_00703 0.67 0.64 0.69 9 4265 18 1 3 14 5
RFA_00765 1.00 1.00 1.00 12 891 1 0 0 1 0
RFA_00770 0.68 0.56 0.83 10 2004 5 0 2 3 8
RFA_00773 0.71 0.56 0.91 10 1942 2 1 0 1 8
RFA_00808 0.75 0.56 1.00 9 2007 0 0 0 0 7
RFA_00816 1.00 1.00 1.00 41 23179 14 0 0 14 0
SRP_00200 0.90 0.92 0.89 33 6866 4 1 3 0 3
SRP_00285 0.95 0.93 0.97 28 3712 1 0 1 0 2
SRP_00308 0.76 0.74 0.77 65 36231 20 2 17 1 23
SRP_00338 0.94 0.94 0.94 34 5424 2 0 2 0 2

^top



Performance of NanoFolder - scored lower in this pairwise comparison

1. Total counts & total scores for NanoFolder

Total Base Pair Counts
Total TP 133
Total TN 95897
Total FP 421
Total FP CONTRA 88
Total FP INCONS 302
Total FP COMP 31
Total FN 242
Total Scores
MCC 0.297
Average MCC ± 95% Confidence Intervals 0.382 ± 0.206
Sensitivity 0.355
Positive Predictive Value 0.254
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
CRW_01535 0.20 0.24 0.17 9 7088 43 7 36 0 28
RFA_00416 0.94 1.00 0.88 15 1468 6 2 0 4 0
RFA_00664 -0.01 0.00 0.00 0 978 13 0 12 1 14
RFA_00703 0.26 0.36 0.19 5 4252 32 5 16 11 9
RFA_00765 0.39 0.42 0.38 5 890 9 2 6 1 7
RFA_00770 -0.01 0.00 0.00 0 1996 24 0 20 4 18
RFA_00773 0.93 1.00 0.86 18 1932 9 3 0 6 0
RFA_00808 0.83 1.00 0.70 16 1993 7 7 0 0 0
RFA_00816 0.00 0.00 0.00 0 23133 89 30 57 2 41
SRP_00200 0.41 0.47 0.35 17 6855 31 4 27 0 19
SRP_00285 0.38 0.43 0.35 13 3704 24 6 18 0 17
SRP_00308 0.14 0.17 0.12 15 36189 112 18 93 1 73
SRP_00338 0.52 0.56 0.49 20 5419 22 4 17 1 16

^top


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