Important work in a very interesting paper. Clement Kent, Howard Hughes Medical Institute 2 February 2015 Liu et al. use the power of whole genome sequencing of multiple haploid drones descending from the same queen to zoom in on where in the genome recombination events are happening. This paper is of special interest to social insect researchers because it suggests that the already high previous estimates of honeybee recombination rates may have been two-fold underestimates. The second major conclusion is about where crossover events occur. The authors show that these are more common in the vicinity of worker-brain expression biased genes and GPCRs involved in signal transduction.The authors have been quite careful in accounting for possible correlated factors which might have produced spurious correlations: especially important, in my view, is their ruling out the possible confounding effects of gene density. I believe the quantitative estimates of numbers of gene conversion events published here will stimulate much further research on the question of whether biased gene conversion due to crossover or non-crossover events contributes to genome patterns of GC content. Competing interests I am an author on several papers whose conclusions are positively cited by this paper, so I am likely to have some a priori bias.