Human meiosis is highly variable. While existing methods have provided useful insight into this variation, they are limited by the number of meiotic phenotypes, gametes or individuals that can be analysed in parallel. Now, a study in Nature describes Sperm-seq, a single-cell sequencing approach that enables genome-wide analysis of multiple meiotic phenotypes in thousands of sperm simultaneously.

Bell et al. used Sperm-seq and accompanying computational methods to sequence and analyse 31,228 sperm cell genomes from 20 young donors (18–38 years old). The team focused on a range of meiotic phenotypes, including recombination rate, crossover location and spacing, and aneuploidy.

Credit: Ed Buziak/Alamy

First, phased parental haplotypes were inferred for each donor genome. Then, crossover events in each sperm genome were identified and located based on switches from one parental haplotype to the other. A total of 813,122 crossovers were detected across all cells. Among donors, recombination rates ranged from 22.2 to 28.1 crossovers per cell (s.d. 1.53). Variation between gametes, which ranged from 17 to 37 events per cell (s.d. 4.23), was higher than between individuals. Notably, the global recombination rate in individuals was indicative of the crossover rate at each chromosome; and, in gametes, the number of crossover events in one-half of the genome reflected the number in the other half. Thus, factors affecting crossover frequency act throughout the nucleus, rather than locally.

Analyses of crossover location and spacing revealed that both parameters varied greatly among donors, with use of proximal (centromeric) crossover zones more variable than distal (telomeric) zones. However, use of distal zones and crossover spacing were found to be inversely correlated with recombination rate. These relationships held true when the analysis was restricted to chromosomes with exactly two crossovers and applied both to donors and to single cells. Taken together, these observations suggest that variation in crossover location and spacing reflects underlying biological variation between people and between cells rather than an indirect effect of the number of crossovers on a chromosome.

Aneuploidy was detected for all chromosomes and donors, ranging from 0.010 to 0.046 aneuploidy events per cell in donors, a 4.5-fold difference that seemed to reflect true inter-individual variation. Crossovers are thought to protect against aneuploidy during maternal meiosis. Analysis of chromosome gains originating in meiosis I (when recombination occurs) suggested that crossovers are similarly protective in sperm, as 36% fewer crossovers were detected than matched, properly segregated chromosomes.

Sperm-seq … enables genome-wide analysis of multiple meiotic phenotypes in thousands of sperm simultaneously

Taken together, the study shows that certain meiotic phenotypes, such as high recombination rate, closely spaced crossovers and proximally located crossovers, co-vary across chromosomes, gametes and donors. The authors propose that inter-cell and inter-individual variation in meiotic chromosome compaction — and therefore the number of crossover events — could explain this covariance.