Table S6

Table S6: Assessment of the performance of RaFAH (Precision, Recall and F1-Score) on Test Set 3 based on the influence of ecosystem source, genome completeness, similarity to genomes on the training set, taxonomic level, and prediction score. Total_Genomes refers to the number of genomes evaluated (regardless of having associated predictions or not) and was used as the total number of genomes used to calculate recall. Total_Valid_Genomes refers to the number of genomes with predicted hosts by RaFAH and by the combined classical approaches and was used as the total number of genomes to calculate precision.