Rapp, Josephine Z; Fernández-Méndez, Mar; Bienhold, Christina; Boetius, Antje (2017): Bacterial and eukaryotic operational taxonomic units (OTU) in sea ice, water and deep-sea sediment samples of the Central Arctic collected during POLARSTERN cruise ARK-XXVII/3 (IceArc) in 2012 [dataset]. PANGAEA, https://doi.org/10.1594/PANGAEA.882580,

---

Supplement to: Rapp, JZ et al. (2018): Effects of ice-algal aggregate export on the connectivity of bacterial core communities in the central Arctic Ocean. Frontiers in Microbiology, 9, https://doi.org/10.3389/fmicb.2018.01035

We aimed to explore the community composition and turnover of eukaryotic and bacterial microorganisms associated with ice-associated and sinking algal aggregates, as well as their similarity to potential source communities of sea ice, water and deep-sea sediments using Illumina tag sequencing. We sampled algae aggregates growing in melt ponds on sea ice, deposited algae aggregates at the seafloor in more than 4000 m water depth, sea ice, upper water column, sediment surface and the gut content of holothurians feeding on the deposits. For Illumina sequencing, the Amplicon libraries of the bacterial V4-V6 region of the 16S rRNA gene and the eukaryotic V4 region of the 18S rRNA gene were generated according to the protocol recommended by Illumina (16S Metagenomic Sequencing Library Preparation, Part # 15044223, Rev. B). For Bacteria we selected the S-D-Bact-0564-a-S-15 and S-*Univ-1100-a-A-15 primer pair based on a primer evaluation by Klindworth et al. (2013, doi:10.1093/nar/gks808) and for Eukaryota the TAReukFWD1 and TAReukREV3 primers (Stoeck et al., 2010; doi:10.1111/j.1365-294X.2009.04480.x). Libraries were sequenced on an Illumina MiSeq platform in 2x300 cycles paired end runs. For Sequence data cleaning & processin, we used cutadapt (v. 1.8.1; Martin, 2011; doi:10.14806/ej.17.1.200) for the removal of primer sequences and a custom awk script to ensure the correct orientation of reads prior to merging. For merging forward and reverse reads we used pear (v. 0.9.5; Zhang et al., 2014; doi:10.1093/bioinformatics/btt593) and trimmed and quality filtered all sequences using trimmomatic (v. 0.32; Bolger et al., 2014; doi:10.1093/bioinformatics/btu170). We reassured correct formatting of the fastq files with bbmap (v. 34.00; B. Bushnell - sourceforge.net/projects/bbmap) before clustering the reads into OTUs by applying a local clustering threshold of d=1 and the fastidious option in swarm (v. 2.1.1; Mahé et al., 2015; doi:10.7717/peerj.1420). Subsequently, we used the SINA aligner (v. 1.2.10; Pruesse et al., 2012; doi:10.1093/bioinformatics/bts252) to align and classify the seed sequence of each OTU with the SILVA SSU database release 123 (Quast et al., 2013; doi:10.1093/nar/gks1219). OTUs that were classified as chloroplasts, mitochondria, archaea, or those that could not be classified at domain level were removed from further analysis. OTUs that were classified as bacteria within the eukaryotic dataset and vice versa, were removed as well. Furthermore, we removed all absolute singletons, OTUs that were only represented by a single sequence across the complete dataset. Filtering and removal of absolute singletons resulted in a final number of 8,869 bacterial and 7,627 eukaryotic OTUs. All further analyses were performed on these processed OTU abundance tables.