Fig. 2 | Nature Communications

Fig. 2

From: RNA secondary structure prediction using an ensemble of two-dimensional deep neural networks and transfer learning

Fig. 2

Performance comparison of SPOT-RNA with 12 other predictors by using PR curve and boxplot on the test set TS1. a Precision-recall curves on the independent test set TS1 by initial training (SPOT-RNA-IT, the green dashed line), direct training (SPOT-RNA-DT, the blue dot-dashed line), and transfer learning (SPOT-RNA, the solid magenta line). Precision and sensitivity results from ten currently used predictors are also shown as labeled with open symbols for the methods accounting for pseudoknots and filled symbols for the methods not accounting for pseudoknots. CONTRAfold and CentroidFold were also shown as curves (Gold and Black) because their methods provide predicted probabilities. b Distribution of F1 score for individual RNAs on the independent test set TS1 given by various methods as labeled. On each box, the central mark indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The outliers are plotted individually by using the “+” symbol.

Back to article page