Skip to main content

Table 1 Correlations in subsequences taken at random from the human genome

From: Genomic DNA from animals shows contrasting strand bias in large and small subsequences

Window length

Correlation coefficient (G - C) v (A - T) unmasked genome

Correlation coefficient (G - C) v (A - T) masked genome

50

+0.140

+0.193

100

+0.176

+0.226

200

+0.208

+0.253

500

+0.216

+0.247

1000

+0.177

+0.232

2000

+0.109

+0.189

5000

+0.012

-0.024

10000

-0.083

-0.199

20000

-0.182

-0.398

50000

-0.342

-0.626

100000

-0.384

-0.739

200000

-0.577

-0.814

500000

-0.672

-0.875

1000000

-0.769

-0.876

2000000

-0.727

-0.868

  1. In each case the sample size is 4000. There is a smooth transition from very small to very large windows. The results for the masked genome are more pronounced than for the unmasked genome.