Dissecting genetic and sex-specific host heterogeneity in pathogen transmission potential

Jonathon A. Siva-Jothy; Pedro F. Vale

contained either a two-way interaction between genetic background and sex or a 245 two-way interaction between genetic background and mating (Table S6) . regressions analysing sex-differences in viral load included DPI (a 3-level factor: 1, 2 253 or 3 days) and an interaction between genetic background and sex (Table S6 ). For 254 analysing the effect of mating in females on viral load, logistic regressions included 255 DPI and an interaction between genetic background and mating (Table S6 ). Logistic 256 regressions of virus shedding used a similar model that also included quantitative 257 viral load as a predictor (Table S6 ). After these logistic regressions, zeroes were 258 removed from all datasets to analyse the subset of positive-values. The GLMs used 259 to analyse these subsets included the same predictors as their corresponding logistic 260 regressions, for viral load: an interaction between genetic background and sex or 261 mating, alongside DPI, with the inclusion of quantitative viral load for virus shedding 262 (Table S6) . 263 264 Due to zero-inflation V was also analysed with a logistic regression followed by a 265 GLM. A logistic regression was used to analyse the proportion of V values that were 266 greater than zero with a two-way interaction between sex and genetic background 267 as predictors (Table S6) . Zero-values of V were then removed from the dataset, and 268 . CC-BY-ND 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/733915 doi: bioRxiv preprint a GLM was used to analyse differences in the size of V, with an interaction between 269 sex and genetic background included as a predictor (Table S6) . The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/733915 doi: bioRxiv preprint