2.4 Microbiological analysis
Genomic DNA was extracted from a 0.5 g frozen coral sample using the TGuide S96 magnetic bead method and the DP812 DNA extraction kit (Tiangen Biotech Co., Ltd., Beijing, China). The V3-V4 variable region of the bacterial 16S rRNA gene was PCR amplified with the forward primer 27 F (5’-AGRGTTTGATYNTGGCTCAG-3’) and the reverse primer 1492 R (5’-TASGGHTACCTTGTTASGACTT-3’) (Zhong et al., 2021). Library preparation was performed using the SMRTbell Template Prep Kit (PacBio, USA), and the PCR reaction was conducted in a Veriti 96-well thermal cycler (Applied Biosystems, USA). Sequencing was carried out on the Sequel II platform after purification.
After obtaining the sequencing data, the raw reads underwent filtering to remove adapter sequences using Trimmomatic (version 0.33) and Cutadapt (version 1.9.1) software (Bolger et al., 2014; Martin, 2011) to obtain clean reads. Paired reads were merged using USEARCH (version 10) (Edgar, 2013), and chimeric sequences were removed using UCHIME (version 8.1). Subsequently, high-quality sequences were obtained for further analysis. The sequences were clustered at a 97 % similarity level using USEARCH (version 10.0) with a default OTU filtering threshold of 0.005 % of the total sequencing count. For ASV analysis, the DADA2 method in QIIME2 (version 2020.6) was employed to denoise the data, with the default filtering threshold of 0.005 % of the total sequencing count.
The obtained high-quality sequences were then annotated against databases such as Silva, Unite, Greengenes, NCBI, Fungene, and MaarjAM for taxonomic classification and functional prediction using BugBase and FAPROTAX. Alpha and beta diversity analyses, community composition analyses, and BugBase and FAPROTAX functional prediction analyses were conducted using BMKCloud (www.biocloud.net). Core OTU analysis was carried out using Venn diagrams to identify shared OTUs among different sample groups, and the composition and relative abundance differences of core OTUs between groups were analyzed using pie charts and heatmaps. Data visualization and figure generation were performed using GraphPad Prism 8.0.2 and RStudio. The data that support the findings of this study are openly available in figshare: https://doi.org/10.6084/m9.figshare.24152901.