Excel workbook containing the source numerical data underlying figures in the manuscript.

Each worksheet corresponds to specific figures and includes all individual data points used to generate the displayed results. Column definitions: For sheets “Figs 1D, 1F, 1G, 2C, S7A and S10”, “Figs 2B and S7BC (scheme A)”, ”Fig 2B (scheme B)”, “S7BC Fig”, and “S10 Fig”: defined as in S2 Data. For sheets “Fig 2A (mean values)”, “S2–S6C Figs (mean)“, and “S9 Fig”, shap_values: mean absolute SHAP values, features. For sheets “Fig 2A (SHAP values)” and “S2–S6C Fig (shap)“, each column corresponds to a model feature. For sheets “S1A Fig”, species; antibiotic: name of the species that the model trained on; clade; R: number of resistant genomes; S: number of susceptible genomes; size: Total number of genomes. For sheet “S1B Fig”, columns are defined as in S1 Data. For sheet “S8 Fig”, species; count: overlapping features in 50% summed SHAP features (%); top_N: the number of 50% summed SHAP features; count_top10: overlapping features in top 10 SHAP features (%). (XLSX)