Skip to main content

Table 2 Number of positives and negatives used for training, tuning, and testing each model

From: Inferring mammalian tissue-specific regulatory conservation by predicting tissue-specific differences in open chromatin

Model Number

Genomes Used in Training

Tissue

Negative Set

Positives (Training, Validation, Test)

Negatives (Training, Validation, Test)

Negatives:Positives (Training, Validation, Test)

1

mm10

Brain

Flanking Regions

21,594, 2416, 4576

35,640, 4018, 7440

1.65:1, 1.66:1, 1.63:1

2

mm10

Brain

OCRs in Other Tissues

21,594, 2416, 4576

427,174, 70,504, 82,172

19.78:1, 29.18:1, 17.96:1

3

mm10

Brain

Large G/C- and Repeat-Matched

21,594, 2416, 4576

175,912, 23,880, 32,008

8.15:1, 9.88:1, 6.99:1

4

mm10

Brain

Small G/C- and Repeat-Matched

21,594, 2416, 4576

35,358, 4776, 6654

1.64:1, 1.98:1, 1.45:1

5

mm10

Brain

Dinucleotide-Shuffled OCRs

21,594, 2416, 4576

215,940, 24,160, 45,760

10:1, 10:1, 10:1

6

mm10

Brain

Non-OCR Orths. of OCRs

21,594, 2416, 4576

25,086, 3456, 4694

1.16:1, 1.43:1, 1.03:1

7

mm10

Liver

Non-OCR Orths. of OCRs

32,498, 4032, 7752

22,890, 2994, 4434

1:1.42, 1:1.35, 1:1.75

8

mm10, hg38, rheMac8, rn6

Brain

Non-OCR Orths. of OCRs

74,688, 9036, 15,266

111,206, 14,650, 19,688

1.49:1, 1.62:1, 1.29:1

9

mm10, rheMac8, rn6

Liver

Non-OCR Orths. of OCRs

81,886, 10,428, 17,688

67,278, 8680, 14,544

1:1.22, 1:1.20, 1:1.22