Spike protein sequences of zoonotic human coronaviruses with metadata and model-predicted host category.

List of zoonotic coronavirus genome sequences sampled from humans used in spike protein machine learning prediction along with genus, GenBank accession ID, and predicted host category, aggregating over all random forests. Coronavirus names refer to taxonomic entities as defined by NCBI taxonomy (whether species or unranked sub-species). Model predictions are given as probabilities associated with each individual category. (CSV)

本数据集包含从人类样本中采集的、用于刺突蛋白（spike protein）机器学习预测的人畜共患冠状病毒基因组序列列表，同时附带其分类学属、GenBank登录号以及经所有随机森林模型聚合得到的预测宿主类别。冠状病毒名称采用NCBI分类学系统所定义的分类学实体，涵盖物种及未分级亚种两类层级。模型预测结果以各独立类别的关联概率值形式给出。（CSV格式）