16S rRNA sequencing and data processing

The gene located in the 16S rRNA V4 region was detected by specific primers, namely, 515F: GTGCCAGCMGCCGCGGTAA and 806R: GGACTACHVGGGTWTCTAAT. The NEBNext^® Ultra™ RNA Library Prep Kit for Illumina^® (E7530 L, NEB) was used to generate sequenced libraries on the Illumina HiSeq platform (Allwegene Technologies Inc., Beijing, China). The raw data were mainly processed using QIIME 2.0, USEARCH (Version 10.0.240), and other R packages mentioned below (21, 22). Trimmomatic was used to filter the nucleotides of poor quality, and reads < 50 nt were removed (parameters: LEADING: 20, TRAILING: 20, MINLEN: 50) (23). FLASH and Pear were used to assemble overlapping read pairs (24, 25). Chimeras were filtered out by UCHIME (26). The clean tags were left after the screening flow above, and they were clustered into operational taxonomic units (OTUs) by the UPARSE algorithm with a sequence similarity no less than 97% (27). Finally, an OTU table was obtained by quantifying the frequency of the OTUs in each sample. Simultaneously, the OTUs were aligned to the SILVA 132 database and assigned taxonomy at the kingdom, phylum, class, order, family, genus, and species levels (28).