Four ITS reference sequence databases (D1–D4) used for the training of the QIIME2 q2-feature-classifier

Four ITS reference sequence databases (D1–D4) were used to train the QIIME2 q2-feature-classifier. Database D1 includes 196,344 fungal ITS reference sequences from the Unite database37, while the D2 database contains 32,013 macrofungi ITS reference sequences from the Unite database. Database D3 contains all reference sequences of the D2 database and 189 macrofungi ITS reference sequences from GenBank (see Table S4 for detailed information). Finally, Database D4 includes all reference sequences of the D3 database and 5,733 macrofungi ITS reference sequences from the RefSeq database (https://www.ncbi.nlm.nih.gov/refseq/). Four QIIME2 q2-feature-classifier datasets (C1–C4) were generated from the four databases (D1-D4) using the fit-classifier-naive-bayes option of QIIME 2 (v2022.2)

本研究选用4个内部转录间隔区（Internal Transcribed Spacer，ITS）参考序列数据库（D1~D4）对QIIME 2的q2-feature-classifier分类器进行训练。其中，数据库D1包含来自Unite数据库（Unite database）的196344条真菌ITS参考序列；数据库D2包含来自Unite数据库的32013条大型真菌ITS参考序列。数据库D3涵盖D2数据库的全部参考序列，以及来自GenBank数据库（GenBank）的189条大型真菌ITS参考序列，详细信息参见附表S4。最终，数据库D4包含D3数据库的全部参考序列，以及来自RefSeq数据库（RefSeq database，链接：https://www.ncbi.nlm.nih.gov/refseq/）的5733条大型真菌ITS参考序列。研究人员基于上述4个数据库（D1~D4），通过QIIME 2（v2022.2版本）的fit-classifier-naive-bayes功能，构建了4个q2-feature-classifier数据集（C1~C4）。