Computational protocol: Temperature shapes coral-algal symbiosis in the South China Sea

Similar protocols

Protocol publication

[…] Strict quality control and sequence filtration were applied to ensure the accuracy of the following analysis. Adaptors, short reads, and low quality reads were first removed by the sequencing company. The PEAR (paired-end read merger) tool was applied to obtain full-length ITS2 rDNA fragments with merging overlapping PE reads to generate ITS2 sequences (, see commands in ). ITS2 tags were demultiplexed into all samples in the QIIME platform by identifying unique barcodes (, see commands in ). We found that the previous ITS2 database included some duplicate sequences, therefore we uploaded the previous ITS2 database to CD-HIT Suite website (, set sequence identity cut-off as 100%, compared both strands, set all other parameters as default, removed the duplicates, merged the annotations and used the results to establish a non-redundant ITS2 database (, ). Next, all of the sequences were aligned to the ITS2 database by BLASTN based on E-value of 1e-5 (see commands in ), the output were then filtered in Microsoft Excel (Microsoft Corp., USA) based on 97% similarity for Symbiodinium subclade identification. For ITS2 rDNA, 97% sequence similarity has been proved to be reliable for Symbiodinium subclade classification and could permit differences of ten base pairs for identification within the same Symbiodinium subclade as a consequence of intra-genomic sequence divergence.Thirteen (sub)dominated Symbiodinium subclades were picked from 131 total aligned Symbiodinium subclades for following analysis. The relative abundances of picked Symbiodinium subclades were >10% in at least one sample or their average relative abundances in all samples were at least around 1% and the filtering results can indicate the original communities well (, ). To profile Symbiodinium communities in different samples, a heat map and scatter plots were created with OriginPro 2015 (OriginLab, USA). To justify significances between Symbiodinium community structures across the sampling sites, a two-way ANOSIM and a permutational multivariate analysis of variance (PERMANOVA) were performed using PRIMER 7 software. To present the relationships of the Symbiodinium community structures from different sampling sites, a nMDS ordination as well as clustering analysis were developed based on the square root-transformed Bray-Curtis similarity matrix of Symbiodinium profiles (Symbiodinium relative abundance) in PRIMER. To test the relationships among environmental factors, Symbiodinium community, and sampling sites, CCA was conducted in Canoco 5. CCA uses ecological data sets to extract synthetic environmental gradients which are the basis for revealing the differential habitat preferences of microbes and has been proved to be reliable when combining spatial and seasonal sampling together, as long as the sample size is granted and the environment conditions can reflect habitat preference of the communities. To identify the phylogenetic relationships among the dominant Symbiodinium subclades found in this study, two phylogenetic trees were constructed based on the Kimura 2-parameter model with uniform rates among sites using Maximum Likelihood in MEGA 6 and Bayesian Inference in Mrbayes, respectively. […]

Pipeline specifications

Software tools PEAR, QIIME, CD-HIT, BLASTN, MEGA, MrBayes
Databases ITS2 Database
Application Phylogenetics
Organisms Galaxea fascicularis