The emergence of resistance to azithromycin complicates treatment of Neisseria gonorrhoeae, the etiologic agent of gonorrhea. Substantial azithromycin resistance remains unexplained after accounting for known resistance mutations. Bacterial genome-wide association studies (GWAS) can identify novel resistance genes but must control for genetic confounders while maintaining power. Here, we show that compared to single-locus GWAS, conducting GWAS conditioned on known resistance mutations reduces the number of false positives and identifies a G70D mutation in the RplD 50S ribosomal protein L4 as significantly associated with increased azithromycin resistance (p-value = 1.08 × 10−11). We experimentally confirm our GWAS results and demonstrate that RplD G70D and other macrolide binding site mutations are prevalent (present in 5.42% of 4850 isolates) and widespread (identified in 21/65 countries across two decades). Overall, our findings demonstrate the utility of conditional associations for improving the performance of microbial GWAS and advance our understanding of the genetic basis of macrolide resistance.

19The emergence of resistance to azithromycin complicates treatment of N. gonorrhoeae, the 20 etiologic agent of gonorrhea. Population genomic analyses of clinical isolates have 21 demonstrated that some azithromycin resistance remains unexplained after accounting for the 22 contributions of known resistance mutations in the 23S rRNA and the MtrCDE efflux pump. 23Bacterial genome-wide association studies (GWAS) offer a promising approach for identifying 24 novel resistance genes but must adequately address the challenge of controlling for genetic 25 confounders while maintaining power to detect variants with lower effect sizes. Compared to a 26 standard univariate GWAS, conducting GWAS conditioned on known resistance mutations with 27 high effect sizes substantially reduced the number of variants that reached genome-wide 28 significance and identified a G70D mutation in the 50S ribosomal protein L4 (encoded by the 29 gene rplD) as significantly associated with increased azithromycin minimum inhibitory 30 concentrations (β = 1.03, 95% CI [0.76, 1.30]). The role and prevalence of these rplD mutations 31 in conferring macrolide resistance in N. gonorrhoeae had been unclear. Here, we experimentally 32 confirmed our GWAS results, identified other resistance-associated mutations in RplD, and 33 showed that in total these RplD binding site mutations are prevalent (present in 5.42% of 4850 34 isolates) and geographically and temporally widespread (identified in 21/65 countries across two 35 decades). Overall, our findings demonstrate the utility of conditional associations for improving 36 the performance of microbial GWAS and advance our understanding of the genetic basis of 37 macrolide resistance in a prevalent multidrug-resistant pathogen.