{# style="max-width:100%; width: 100px;">#}
Home / Result /
Custom selection
RNA type
circle_RNA
RBP type
ALKBH5
Feature generation methods
bert_3mer
Models to train and compare
CNN
RNN
ResNet
Model prediction results
Performance of deep learning models
Model AUC ACC Recall F1-Score Precision Specificity AP MCC
CNN 0.999 0.987 0.978 0.986 0.995 0.995 0.999 0.974
RNN 0.997 0.984 0.978 0.984 0.989 0.990 0.997 0.969
ResNet-1D 0.999 0.984 0.978 0.984 0.989 0.990 0.999 0.969
User input data analysis
Density distribution of the prediction confidence by different models
Performance metrics visualization

pointplot.png

roc_curve.png

precision_recall_curve.png

upset_plot.png

barplot.png

boxplot.png

det_curve.png

violinplot.png

Id Methods Header Line RBP binding site probability Bind(threshold: 0.8)
1 CNN >chr1,+,10068791,10068891; class:0 0.000 False
2 CNN >chr1,+,10468105,10468205; class:1 1.000 True
3 CNN >chr1,+,10475688,10475788; class:0 1.000 True
4 RNN >chr1,+,10068791,10068891; class:0 0.391 False
5 RNN >chr1,+,10468105,10468205; class:1 0.951 True
6 RNN >chr1,+,10475688,10475788; class:0 0.993 True
7 ResNet-1D >chr1,+,10068791,10068891; class:0 0.000 False
8 ResNet-1D >chr1,+,10468105,10468205; class:1 1.000 True
9 ResNet-1D >chr1,+,10475688,10475788; class:0 1.000 True