Supplementary material for: Using machine learning to attribute the source of antimicrobial resistant campylobacteriosis in the United States

<b>Background </b>Combined pathogen genomic surveillance with advanced bioinformatics analyses has the potential to inform public health risk and targeted interventions. In this study, we analyse the two most common pathogenic <i>Campylobacter</i> species in human gastrointestinal infection. These enteric bacteria are ubiquitous in the gut of birds and mammals, which can commonly infect humans via consumption of contaminated food. Rising incidence and antimicrobial resistance (AMR) are a major global concern and there is an urgent need to quantify the main routes to human infection. We test the utility of machine learning methods for probabilistic assignment of genome sequenced campylobacteriosis isolates to possible sources in the United States between 2009 and 2019.<b>Methods</b> As part of routine US national surveillance (2009 through 2019), 8,856 <i>Campylobacter </i>isolate genomes were sequenced from human infections and 16,703 from possible sources. Targeting genetic variation associated with host adaptation, we used machine learning and probabilistic models to attribute the source of human infections and estimate the relative importance of different disease reservoirs.<b>Findings</b> Probabilistic attribution identified poultry as the primary infection source of human clinical isolates, responsible for an estimated 68% of cases. Most of the remaining clinical isolates were attributed to cattle (28%), with only a small contribution from wild bird (3%) and pork sources (1%). An increase in fluoroquinolone and aminoglycoside resistant isolates underpinned increased multidrug resistance in human infection cases, that could be attributed to chicken sources.<b>Interpretation </b>National-scale surveillance and quantification of the relative contribution of infection sources can guide policy. Our study suggests that the greatest reductions in human campylobacteriosis in the US will come from interventions that focus on poultry, which may also reduce the spread of AMR.