Lncident: A Tool for Rapid Identification of Long Noncoding RNAs Utilizing Sequence Intrinsic Composition and Open Reading Frame Information
Table 1
The performances on human dataset 2.
Tools
Sensitivity
Specificity
Accuracy
-measure
MCC
Kappa
CPC.local
0.6613
1
0.8306
0.7961
0.7028
0.6612
CPC.web
0.6625
0.9992
0.8309
0.7966
0.7028
0.6618
CPAT.local
0.9333
0.9802
0.9568
0.9557
0.9145
0.9135
CPAT.web
0.9535
0.9673
0.9604
0.9601
0.9208
0.9208
CNCI
0.9702
0.9157
0.9430
0.9445
0.8873
0.8860
PLEK
0.9952
0.8918
0.9435
0.9463
0.8918
0.8870
CPAT.train
0.9160
0.9848
0.9504
0.9486
0.9029
0.9008
PLEK.train
0.7622
0.9507
0.8565
0.8416
0.7260
0.7130
Lncident
0.9535
0.9795
0.9665
0.9661
0.9333
0.9330
Lncident displayed a satisfying overall performance. CPC and CPAT were tested on stand-alone version and web server. suffix of “train” means CPAT and PLEK with the new-trained model.