Skip to main content

Table 3 Signatures detected in top 20 ranked features (Human)

From: Identification of long non-coding transcripts with feature selection: a comparative study

Signatue #

Algorithm groups

BASIC

CONS

NUCLEO

ORF

REPS

AUPR (AUC)

1

IG, RFS,

TxExLenAvg,

ph100m,

AA, AAT, AT,

KOZAK,

DNA.TcMar.Tigger,

0.69 (0.94)

 

RF,

TxLen,

ph20m,

ATA, CA, CC,

OrfProp

LINE.L1,

 
 

EFmn

TxNex

ph20mn,

CCG, CG,

 

LTR.ERV1,

 
   

ph20mx,

CGA, CGT,

 

LTR.ERVL,

 
   

py100m,

FickScore, GC,

 

LTR.ERVL.MaLR,

 
   

py100mx,

GG, GT, GTG,

 

SINE.Alu,

 
   

py20m

TA, TAT, TCG,

 

SINE.MIR

 
    

TT, TTA

   

2

GR

TxExLenAvg

ph100m,

ATC, ATG, CA,

 

DNA.DNA,

0.55 (0.92)

   

ph20m,

CAC

 

DNA.hAT.Blackjack,

 
   

ph20mx,

  

DNA.MULE.MuDR,

 
   

py100m,

  

DNA.PiggyBac,

 
   

py100mx,

  

DNA.TcMar.Tc2,

 
   

py20m

  

LINE.Penelope,

 
      

LTR.LTR,

 
      

RC.Helitron,

 
      

SINE.MIR

 

3

GFS

TxExLenAvg,

ph100m,

AA, ACC, CA,

KOZAK

LINE.Penelope

0.67 (0.94)

  

TxLen,

ph20mx,

CAG, CTA,

   
  

TxNex

py100m,

FickScore,

   
   

py20m

GAT, GT,

   
    

TAC, TAT,

   
    

TGG

   

4

LR, EN

TxLen,

ph100m,

AA, AAT, ACA,

KOZAK

 

0.66 (0.94)

  

TxNex

ph20m,

ACT, CA,

   
   

ph20mx,

CAA, CAC,

   
   

py100m,

CG, CGA,

   
   

py100mx

FickScore, GG,

   
    

GT, GTG,

   
    

TAC, TCT,

   
    

TGA, TGG

   

5

5 WT

TxExLenAvg,

 

AAC, AAG,

  

0.66 (0.94)

  

TxNex

 

AC, ACA,

   
    

ACC, ACG,

   
    

ACT, AGA,

   
    

AGC, AGT,

   
    

ATA, CA, CT,

   
    

GA, GT, TA,

   
    

TC, TG