Structured, Inbred and Plastic: the Genome and Population Genetics of the weed False cleavers (Galium spurium)

Abstract

False cleavers (Galium spurium L.) is an aggressive weed from the Rubiaceae. Here we assemble a chromosome scale draft of its genome, laying the foundations for determining the genetic basis of auxinic herbicide resistance and for systematic research into its polyphyletic genus. We use the genome to examine the population genetics in material from the Canadian Prairies and, in concert with a common greenhouse experiment, to examine whether the phenotypic variation observed in the field results primarily from genetic or environmental factors. The genome assembly covers approximately 85% of G. spurium’s expected 360Mbp genome size with 94% of BUSCO genes complete and most single copy (89%). Approximately 37% of the genome is repetitive elements and 35,540 genes were annotated using RNA-Seq data, including 100 homologs for genes involved or, potentially involved in, herbicide resistance. The genome shows strong synteny with other members of the Rubiaceae including smooth bedstraw (Cruciata laevipes Opiz) and robusta coffee [Coffea canephora (Pierre ex Froehner]. Double digested RADseq data for the 19 populations from the Canadian Prairies indicated that G. spurium has high levels of population structure (F_ST = 0.54) and inbreeding (F_IS =0.86) with low levels of hetrozygosity (H_O = 0.02) and nucleotide diversity (π = 0.0003). Variation in flowering time and seed weight largely overlapped among populations grown in the greenhouse. A redundancy analysis investigating genotype-phenotype associations showed few associations between SNP variation and these characteristics. In contrast, the majority of SNPs under selection were associated with mericarp hook density. This suggests that for most traits, environmental variation rather than genetic variation likely underlies phenotypic differences observed in the field. Several genes of interest including several homologs involved in the assembly of the Skp1-Culliun-F-Box IR1/AFB E3 ubiquitin ligase complex (e.g. CAND1, ECR1) are located in areas of the genome with evidence of selection and are targets for further investigation.