Skip to main content
. 2018 Mar 28;14(3):e1007254. doi: 10.1371/journal.pgen.1007254

Table 7. Correlation between the GC-corrected density of DNMs and various genomic variables.

Individual correlations Multiple regression
Francioli Wong Jonsson Francioli Wong Jonsson
Male recombination rate 0.092*** 0.18*** 0.15*** 0.099*** 0.15***
Female recombination rate 0.10*** 0.16*** 0.074*** 0.078** 0.072**
H3K4me1 -0.006 0.015 -0.028 0.13** 0.12**
H3K4me3 -0.096*** -0.12*** -0.090** -0.091** -0.18***
H3K27me3 -0.040 -0.023 -0.036
H3K27ac -0.047* -0.037 -0.044* 0.11**
Transcription rate -0.050* -0.094*** -0.029
H3K4me1PB -0.051* -0.015 -0.080*** 0.055* -0.071*
H3K9me3PB 0.022 0.004 0.032
Nucleosome occupancy 0.020 0.040 -0.001
DNAse hypersensitivity -0.031 -0.024 -0.026
Replication time -0.090*** -0.079*** -0.10*** -0.13*** -0.087* -0.14**
PC1 -0.040 -0.026 -0.048*
PC2 0.14*** 0.23*** 0.15***
PC3 -0.012 -0.010 0.000
PC4 0.025 -0.005 0.084***

Also given are standardised slopes from a multiple regression with forward parameter selection (p<0.05 for inclusion). The density of DNMs was corrected for GC-content by regressing DNM density against GC-content and the squared GC-content and taking the residuals.

* p < 0.05

** p < 0.01

*** p < 0.001