Skip to main content
Fig. 2 | Epigenetics & Chromatin

Fig. 2

From: Affinity-seq detects genome-wide PRDM9 binding sites and reveals the impact of prior chromatin modifications on mammalian recombination hotspot usage

Fig. 2

Hotspot usage and chromatin marks. a Detection of historical hotspot usage using CAST/EiJ genome for SNP calling. SNPs from the CAST/EiJ genome were obtained from the Wellcome Trust Sanger Institute’s Mouse Genomes Project. SNP densities are significantly increased at the PRDM9 binding sites of in vivo hotspots indicative of biased gene conversion at active hotspots, but only slightly so for sites that were not detectably active in vivo. Regions spanning 2 kb are centered and oriented based on the PRDM9 binding sites identified by Affinity-seq; SNP frequency for each nucleotide position is expressed as ratio of increase over the regional average. b Influence of genomic features on hotspot usage. The boundaries of H3K9me3 (+) or (−) and H3K9me2 (+) or (−) regions were determined using the corresponding ChIP-seq data and rseg. cLADs represent constitutive Lamin B-associated domains, ciLADS represent regions never associated with Lamin B domains. Gene expression is determined by ChIP-seq in 12-dpp spermatocytes. The bars show the fraction of in vivo used Affinity-seq binding sites relative to the total number of Affinity-seq binding sites in each category. The sites in each group are counted as belonging to a category if they cross into or are encapsulated within its borders. The fractions for various genomic features representative of opened and closed chromatin environments reveal a consistent reduction of in vivo usage in heterochromatic regions. Highly expressed genes represent the top quartile of the genes expressed in testes of 12-dpp mice. The bars show the fraction of in vivo used Affinity-seq binding sites relative to the total number of Affinity-seq binding sites in each category. Standard error bars are calculated from the Poisson distribution. c Distribution of H3K9me2, H3K9me3, and hotspot-deficient regions along mouse Chromosome 1. Top to bottom in each chromosome, first panel, H3K9me3 enriched (up) or depleted (down) regions; second panel, H3K9me2 enriched and depleted regions; third panel, regions containing Affinity-seq but deficient in H3K4me3/DMC1 peaks (in vivo Affinity-seq sites) in dispersed regions identified by Rscan are represented as red rectangles, those associated with assembly gaps removed; fourth panel, recombination-deficient regions in Collaborative Cross mice [30], red rectangles; fifth panel, alignability track from the UCSC browser showing assembly gaps and low mappability regions

Back to article page