Fig. 5.

COs and NCOs are depleted in asymmetric hotspots in both sexes. a Human-controlled DMC1 hotspots were separated into three bins (asymmetric, intermediate, and symmetric) according to symmetry, so that each bin contains the same number of predicted events according to DMC1 enrichment (Supplementary Note 8). Grey bars show the DMC1-predicted expected fraction of events in each bin. The four coloured bars (vertical lines: 95% CIs) show the observed fraction of (sampled) F5 de novo events: COs, NCOs, and paternal or maternal recombination events. b As a, except predicted events were defined using H3K4me3. c, d Expected vs observed initiation biases at symmetric (c) and asymmetric (d) hotspots, as defined by H3K4me3 symmetry. If asymmetry-related recombination deficiency occurred equally on both homologues, then the fraction of DMC1 reads and the fraction of NCO and CO initiations on the less-bound homologue would be predicted to closely match its fraction of H3K4me3, as in the symmetric case (c). Error bars are 95% bootstrap CIs