Skip to main content

Table 1 Locus information for regions and prediction performances

From: Genetic sequence-based prediction of long-range chromatin interactions suggests a potential role of short tandem repeat sequences in genome organization

R

TCR

#TP

#NP

Test AUC

R

TCR

#TP

#NP

Test AUC

    

A

B

C

D

    

A

B

C

D

GM12878

 

0

chr7:115847372-115857098

63

226

0.7417

0.7538

0.8979

0.9042

5

chr7:90224881-90229046

34

122

0.8078

0.8307

0.9221

0.9118

1

chr7:115890993-115892266

56

234

0.7141

0.7341

0.8876

0.8960

6

chr7:116434729-116454408

33

292

0.7785

0.7787

0.7308

0.7036

2

chr7:115861595-115870968

52

252

0.7346

0.7763

0.9152

0.9376

7

chr7:90337078-90341001

32

158

0.8163

0.8275

0.9286

0.9324

3

chr5:131722317-131724751

39

91

0.6122

0.6547

0.8666

0.8286

8

chr22:32162110-32166713

31

127

0.7779

0.7832

0.7789

0.7738

4

chr5:131892428-131895867

34

80

0.5971

0.6343

0.8889

0.8543

9

chr21:34819525-34821921

30

201

0.6704

0.6694

0.7157

0.6901

K562

 

0

chr22:32764253-32784733

46

105

0.8163

0.8121

0.9308

0.9382

5

chr7:89787744-89795672

35

118

0.8546

0.8648

0.8566

0.8727

1

chr22:32920308-32927723

45

109

0.6808

0.7242

0.7744

0.7972

6

chrX:153625659-153635385

34

46

0.8501

0.8495

0.8044

0.8184

2

chr22:32012966-32043914

42

104

0.7145

0.7324

0.8378

0.8599

7

chr22:32170492-32188129

32

97

0.7456

0.7146

0.8003

0.8228

3

chr21:35242603-35256847

39

150

0.7321

0.725

0.7251

0.7407

8

chr22:32740683-32750950

32

112

0.7167

0.7582

0.8836

0.9166

4

chr7:115847372-115857098

37

238

0.7521

0.7756

0.7765

0.7908

9

chr11:5721056-5732713

31

85

0.671

0.76

0.7345

0.7545

HeLa-S3

 

0

chr7:115847372-115857098

98

207

0.6914

0.7111

0.8007

0.8228

5

chr7:115861595-115870968

40

284

0.6624

0.732

0.8964

0.9114

1

chr7:116434729-116454408

71

211

0.73

0.7674

0.8573

0.8738

6

chr22:32170492-32188129

40

102

0.677

0.755

0.8245

0.8590

2

chr22:32920308-32927723

53

109

0.644

0.6369

0.7338

0.7091

7

chr22:32053085-32061138

37

115

0.6018

0.6420

0.7886

0.7991

3

chr7:115890993-115892266

50

243

0.6817

0.7225

0.907

0.9162

8

chr22:33262063-33266567

37

112

0.5634

0.6564

0.8449

0.8491

4

chr7:89787744-89795672

49

108

0.8108

0.8007

0.8005

0.8084

9

chr21:34750664-34761738

37

147

0.7194

0.7294

0.7053

0.7273

  1. #TruePeaks (#TP) and #NonPeaks (#NP) for all the studied genomic regions (column ‘R’) for the three cell lines (GM12878, K562 and HeLa-S3). Columns marked ‘A’, ‘B’, ‘C’ and ‘D’ show the mean test AUC values with oligomer length 3 and 5 respectively for two settings: Individual tasks (‘A’ and ‘B’) and Multiple tasks (‘C’ and ‘D’). Refer “Pipeline for predicting long-range chromatin interactions”, “Prediction of long-range chromatin interactions is possible from the sequence alone using non-linear SVMs” and “Multitask learning (MTL) helps mitigate issue of having too few interacting partners per locus” sections for more information