Whole genome sequences used in analysis with metadata and model-predicted host category.

List of non-zoonotic coronavirus genome sequences used in whole genome machine learning analysis along with genus, GenBank accession ID, metadata-derived host category and predicted host category from random forest when respective coronavirus was held-out during validation. Coronavirus names refer to taxonomic entities as defined by NCBI taxonomy (whether species or unranked sub-species). Model predictions are given as single category with highest probability, along with probabilities associated with each individual category. (CSV)