Classical models suggest that recombination rates on sex chromosomes evolve in a stepwise manner to localize sexually antagonistic variants in the sex in which they are beneficial, thereby lowering rates of recombination between X and Y chromosomes. However, it is also possible that sex chromosome formation occurs in regions with pre-existing recombination suppression. To evaluate these possibilities, we constructed linkage maps and a chromosome-scale genome assembly for the dioecious plant Rumex hastatulus. This species has a polymorphic karyotype with a young neo-sex chromosome, resulting from a Robertsonian fusion between the X chromosome and an autosome, in part of its geographical range. We identified the shared and neo-sex chromosome using comparative genetic maps of the two cytotypes. We found that sex-linked regions of both the ancestral and the neo-sex chromosome are embedded in large regions of low recombination. Furthermore, our comparison of the recombination landscape of the neo-sex chromosome to its autosomal homologue indicates that low recombination rates preceded sex linkage. These patterns are not unique to the sex chromosomes; all chromosomes were characterized by massive regions of suppressed recombination spanning most of each chromosome. This represents an extreme case of the periphery-biased recombination seen in other systems with large chromosomes. Across all chromosomes, gene and repetitive sequence density correlated with recombination rate, with patterns of variation differing between repetitive element type. Our findings suggest that ancestrally low rates of recombination may facilitate the formation and subsequent evolution of heteromorphic sex chromosomes.

The data here include a whole-genome assembly from Dovetail Genomics of a Rumex hastatulus male from the XX/XY-cytotype population, an .agp file relating that assembly to a linkage map, a pseudomolecule assembly, a repeat-filtered gene annotation, and a repeat annotation.

### Genome assembly

    ### Primary assembly 
        hastate_28Sep2018_nbKXS.fasta # Whole-genome assembly received from Dovetail Genomics of a TX (XX/XY-cytotype) male

        no_eq_no_sc_DEGAPPEDhastate_28Sep2018_nbKXS.fasta # The genome assembly received from Dovetail Genomics with special characters (line breaks within scaffolds, and semicolons and equals signs in scaffold headers) removed, because some programs don't expect those characters in fastas
    
    ### Secondary assembly
    
        CHRR_genome.agp # The .agp file relating our linkage groups to the scaffolds in the primary assembly. LG identities are as follows: A1 = L.7, A2 = L.8, A3 = L.5, A4 = L.3, X = L.10
        
        CHRR_integrated.fa # Pseudomolecules assembled from the linkage groups
        

### Genome annotation
    
    ### Gene annotations
        
        goodGenes.chrom.gff # Current Augustus genome annotation, positioned relative to the linkage maps
    
    ### Repeat annotations
        
        LG.repeats.gff # Repeat annotation, positioned relative to the linkage maps