Computational protocol: Close Link Between Harmful Cyanobacterial Dominance and Associated Bacterioplankton in a Tropical Eutrophic Reservoir

Similar protocols

Protocol publication

[…] Sequencing was performed in a MiSeq platform (Illumina) using the MiSeq Reagent Kit v3 (2 Ɨ 300 base pairs) according to the manufacturer's instructions. Files were recovered (.fastq), and paired-end reads were joined using mothur v.1.35.1 (Schloss et al., ). Sequences are available for download via the NCBI short read archive under BioProject PRJNA406945. The following criteria were used to eliminate low-quality reads: average quality (window size = 50) <30, length 460 base pairs, presence of ambiguous characters (ā€œNā€), homopolymer <8. The remaining reads were aligned using the SILVA database, trimmed and filtered. Then, sequences were preclustered with diff = 4. Chimeras were detected using UCHIME (Edgar et al., ) and excluded. Taxonomic classification was carried out using the RDP database (Release 11) with a confidence threshold of 80%. Sequences not assigned as Bacteria or classified as Chloroplast or Mitochondria were discharged. Singletons and doubletons were removed, and the number of sequences in the 22 samples was normalized to the same number of sequences. Sequences were then used as input to generate a distance matrix and clustered into operational taxonomic units (OTUs) at the sequence similarity cutoff of 97%. Species richness and Shannon diversity index were calculated in mothur. Taxonomic assignment of OTUs was performed using Greengenes (version 13_5).The OTU relative abundances in the samples were used to generate an ordination plot by nMDS (non-metric multidimensional scaling) based on Bray-Curtis similarity coefficients. The limnological parameters and cyanobacterial cell counts with significant differences between the periods (t-test, p < 0.001) were plotted together with the nMDS. Statistical significance of sample grouping was tested with PERMANOVA. These analyses were performed on package Past3 (Hammer et al., ). To identify the major OTU contributors to grouping differentiation (periods), we used a similarity percentage analysis (SIMPER) (Clarke, ). Spearman correlation was used to test the degree of association among limnological variables with cyanobacterial cell counts and cyanobacterial OTUs and to test the association between specific cyanobacterial OTUs and heterotrophic bacterial OTUs (considering only those that contributed at least 0.2% to the total of sequences). We considered relevant correlations those with p < 0.001 and r > 0.6. The visualization of these correlations was made with the Cytoscape package version 3.4.0 (available at: using the plugin CoNet. The r-value (>0.6) was selected to support the generation of an edge-weighted spring-embedded network (Assenov et al., ). The network included the classified OTUs which contributed with at least 0.2% of the total. […]

Pipeline specifications

Software tools mothur, UCHIME, CoNet
Application 16S rRNA-seq analysis
Organisms Escherichia coli, Microcystis aeruginosa, Bacteria, Bacteroidetes