Computational protocol: Determining Microeukaryotic Plankton Community around Xiamen Island, Southeast China, Using Illumina MiSeq and PCR-DGGE Techniques

Similar protocols

Protocol publication

[…] Total DNA from twelve sites was sent to Novogene Co., Ltd. Beijing for Illumina MiSeq sequencing using a paired-end 150 bp sequence read run with the Miseq Reagent Kit v3. A set of primers were applied to amplify hypervariable V9 region in eukaryotic 18S rDNA. In this study, the forward and reverse primers were 1380F and 1510R []. Twelve DNA samples were independently amplified in triplicated 25 μL reactions including 5 min denaturation at 94°C, then 25 cycles of 30 seconds at 94°C, 30 seconds at 50°C and 30 seconds at 72°C. The final amplification was 7 min extension at 72°C. In total 1 × PCR buffer, 0.625 U Ex-Taq polymerase, 2.5 mM dNTPs, 20 ng target DNA and 10 μM each primer were included in 25 μL reaction system. The FLASH [] was used to merge reads pairs from DNA fragments. Quality-checked sequences were analyzed by QIIME v1.7.0 (Quantitative Insights Into Microbial Ecology) [], where data procedure were as follows: a) maximum of continuous low-quality base is three; b) minimum of continuous high-quality base is 75% of total read length; c) no ambiguous (N) character exists in sequences; d) last quality score is three. Standardized sequence number of twelve samples was selected randomly []. After that, we used UPARSE [] to pick OTUs (operational taxonomic units) by constructing OTU table. All reads were assigned into OTUs at 97% similarity threshold by uclust v1.2.22q []. One representative sequence of each OTU was selected and matched with SILVA [] sequences by RDP [] classifier to identify phylogenetic affiliation [ – ]. Chimera and singleton sequences were discarded prior to further analysis []. [...] DGGE profiles were analyzed by Quantity One (BioRad, Hercules, CA, USA). The software can automatically detect the bands, besides the results can be manually checked. DGGE bands profile and Miseq sequencing standardized OTUs were transformed into binary code, where ‘1’ and ‘0’ indicated ‘presence’ and ‘absence’, respectively. Two Bray-Curtis similarity matrices based on DGGE and Illumina data were constructed, respectively. Then the non-metric multidimensional scaling analysis (MDS) was employed for detecting patterns in microeukaryotic communities among north, east, south and west stations. Significant difference (P < 0.01) between groups was assessed by analysis of similarities (ANOSIM). The global R statistic ranges from 0 to 1 represents separation degree between site groups, and no separation is indicated by R = 0, while R = 1 indicates complete separation [].All physicochemical parameters except pH were square-root transformed for improving homoscedasticity and normality []. After that, preliminary DCA (detrended correspondence analysis) on biological data was applied to decide whether linear or unimodal ordination methods should be used. The DCA result based on DGGE data showed the longest gradient length was less than 3.0, while DCA on Miseq sequencing OTU data revealed that longest gradient length was between 3.0 and 4.0, thereby implying most taxa displayed linear responses to environmental factors. Therefore, we chose redundancy analysis (RDA) to examine the relationship between microeukaryotic community and environmental factors.Shannon-Wiener index (H’) was calculated based on microeukaryotic DGGE profiles and Miseq sequencing OTUs. The H’ was calculated using the following equation: H’ = -ΣP ilnP i. The term P i was determined by the equation: P i = n i/N, where n i is the number of ith band or OTU in a sample and N is the sum of the band or sequence numbers. Both Scheffe’s F multiple-comparison test and ANOVA (analysis of variance) were employed to detect differences among twelve stations. Statistical analyses were run in STATISTICA 6.0, PRIMER 5.0, ORIGIN 8.0, CANOCO 4.5 and the R software packages.The Z score was calculated for the relative abundance of microbial eukaryotes from Miseq sequencing heatmap. The Z score was determined using the equation: Z=(xi−x¯)/SD. Where xi is relative abundance of a certain microeukaryote, and x¯ is average relative abundance in all sampling sites. […]

Pipeline specifications

Software tools QIIME, UPARSE, UCLUST, RDP Classifier, Statistica
Applications Miscellaneous, Phylogenetics, 16S rRNA-seq analysis
Diseases Pulmonary Fibrosis