2.4.2 Sequencing data processing
Raw sequence data were quality-filtered, and chimera was checked using the QIIME software (version 1.8.0) to remove reads containing more than 10% unknown nucleotides and reads where fewer than 50% of all bases had quality values (Q-values) >20 (Caporaso et al ., 2010). Operational taxonomic units (OTUs) were clustered with a sequence threshold of 97% similarity by UPARSE39, and representative sequences of OTUs were picked up simultaneously. The tag sequence with the highest abundance within each cluster was selected as the representative sequence. The taxonomic assignment of 16S rRNA sequences was determined using the bacterial SSUrRNA reference database with the Mothur and SILVA (http://www.arb-silva.de/) classifier, and ITS sequences was determined using the Unite reference database (http://unite. ut.ee/index.php) with the Ribosomal Database Project (RDP) classifier at a 97% level (Edgaret al ., 2011).