*2.4. Data Analysis*

Low-quality regions of the sequences were removed by Cutadapt (version 1.9.1) [27], and then the barcode and primer sequences were cut off to obtain the raw reads. Chimeric sequences were removed using USEARCH (version 8.1.1861) [28], and clean reads were clustered into operational taxonomic units (OTUs) at a 97% similarity level by UPARSE (version 7.1) [29]. OTUs were annotated according to the UNITE database ranging from kingdom to species level [30] and verified via manual search. Five metrics, Shannon, Chao 1, ACE, Simpson, and Good's coverage, were calculated by QIIME (version 1.9.1) [31] to assess alpha diversity. Statistical differences between MOR and AOF were examined by analysis of similarity (ANOSIM). Principal coordinate analysis (PCoA) on the basis of the Bray–Curtis distance matrix was applied to estimate the difference in the fungal community of samples from different species. The linear discriminant analysis effect

size (LEfSe) algorithm (LDA score = 4.0) was performed to distinguish the differentially abundant taxa between two groups [32]. Samples were hierarchically clustered by the unweighted pair group method with arithmetic mean (UPGMA) based on unweighted UniFrac distances. R tools (version 2.15.3) were applied to plot the rarefaction curves, heat map, and Venn diagram.
