Population genomic identification of drug-resistance elements/Information on in vitro SbV-resistance was available for 50/191 Core 191 isolates, from which 25 were sensitive and 25 resistant (Supplementary file 1). Links between genetic diversity (SNP, indel, CNV and somy) and in vitro SbV-resistance were assessed using the Fisher Exact test (FET), Mann Whitney U-tests (MWU) and odds ratios (ORs), implemented on 103 CNVs and 17 indels (in 14 genes) as well as 2,392 phased SNPs genotypes. SNPs were assigned to the 5’ and 3’ UTR if they were within 1 kb of the start or end of the gene (respectively). To counter bias associated with the small sample size, FET and MWU were used initially. For the FET, variants were defined as discrete variables: SNPs as 0, 1 or 2 non-reference alleles, and small indels as the diploid number of inserted or deleted basepairs. For the MWU, mutations were considered as a continuous variable such that the somy state was the haploid chromosome state, and CNVs were the haploid copy number times the somy state. The null hypothesis was that there were no significant genetic differences between SbV-R and SbV-S strains (subject to p < 0.01). The FETs and MWU were limited by the partial association of different mutations with the phenotypes, so we examined ORs of the derived alleles segregating in multiple ISC populations with 6+ non-reference alleles for which the absolute difference in SbV-R and SbV-S allele frequencies > 0.1 using the log-scaled EC50 values. We compared the log-scaled EC50 values of each allele pair using t-tests.