Skip to main content

Table 3 A: DBN results for the RNA-Seq data set. B: SVM results for the RNA-Seq data set

From: Comparing biological information contained in mRNA and non-coding RNAs for classification of lung cancer patients

A.
  DBNDBN and SVM
ModelArchitectureA %TPR %TNR %A %TPR %TNR %
DBN + BpropA-500-250-100-195.48±0.2598.06±0.4092.90±0.3694.13±0.2797.58±0.2791.94±0.34
DBN + BpropA-100-195.65±0.1397.90±0.2593.39±0.1694.19±0.1195.16±0.5093.23±0.22
DBN + RpropA-5-10-195.73±0.1798.39±0.5093.06±0.3495.89±0.0898.39±0.5093.39±0.16
DBN + RpropA-50-195.16±0.5098.39±0.5091.94±0.5095.97±0.5098.39±0.5093.55±0.50
Task: AC vs N; non-coding, A=3124
DBN + BpropA-2000-1000-500-196.77±0.50100±0.5093.55±0.5095.16±0.5096.94±0.1693.39±0.16
DBN + BpropA-100-195.97±0.50100±0.5091.94±0.5095.48±0.1397.42±0.2693.55±0.50
DBN + RpropA-5-10-196.05±0.2298.39±0.5093.71±0.4596.45±0.1398.06±0.2294.84±0.22
DBN + RpropA-50-194.35±0.50100±0.5088.71±0.5095.48±0.1398.39±0.5092.58±0.26
B.
  RadialLinear
DataFeaturesA %TPR %TNR %A %TPR %TNR %
Protein-coding1236093.66±0.8192.66±0.8794.66±0.9193.00±0.8990.00±0.8996.00±0.85
Non-coding312491.41±0.9788.00±0.9994.83±0.5693.91±0.8990.50±0.7397.33±0.72
  1. The best results are shown in bold