Table 2.
Summary of parameter estimates and significance levels of the multivariate generalized linear regression analysis for CpG transition rate
| Dataset A | Dataset B | Dataset C | ||||
|---|---|---|---|---|---|---|
| Estimate | -log10(p) | Estimate | -log10(p) | Estimate | -log10(p) | |
| Divergence | 0.1832 | > 15.70 | 0.1574 | > 15.70 | 0.1722 | > 15.70 |
| Methylation level | 0.0385 | 9.50 | 0.1047 | > 15.70 | 0.3454 | > 15.70 |
| GC content | -0.1036 | > 15.70 | -0.1382 | > 15.70 | -0.1849 | > 15.70 |
| Transcription level | -0.0145 | 1.95 | -0.0094 | 3.01 | -0.0114 | 1.46 |
| Female recombination rate | 0.0031 | 0.22 | 0 | 0 | 0.0158 | 2.29 |
| Male recombination rate | -0.0054 | 0.43 | -0.0084 | 2.29 | -0.0167 | 2.42 |
| Explained deviance | 1,243 | 5,310 | 8,038 | |||
| Residual deviance | 5,894 | 10,496 | 3,451 | |||
| na | 13,038 | 21,636 | 3,871 | |||
| AICb | 37,300 | 78,030 | 17,172 | |||
Dataset A contains no CpG islands (CGIs) or DNase I hypersensitive sites (DHSs)), dataset B contains DHSs but no CGIs, and dataset C contains both CGIs and DHSs. aSample size. bAkaike Information Criterion.