Dataset: HRR Metabarcoding
Deployment: PS1809

Fastq files for all 13 station samples were quality checked using FastQC. Illumina paired end reads (2x300 bp) were processed in mothur v1.39.0 (Schloss et al. 2009). Contigs were assembled and pre-cleaned processed for homopolymers and ambiguities. These sequences were then screened for chimera using UCHIME in denovo mode (Edgar et al. 2011). Sequences were de-noised by pre-clustering at 1 bp per 100 bp and generate unique sequence for taxonomic annotation and characterization. Sequences less than three were eluted out in our study and rest OTUs were characterized using BLAST search using PR2 database v4.12.0. The BLAST analysis used a assignment approach with similarity was ≥ 90% and query coverage was ≥ 70% against the reference sequence. Any OTU that did not compile with this criterion was not used in this study. The following thresholds for identity with BLAST results were used for taxonomic assignment clustering: species (97%), genus (94%), family (93%), class (92%) and order (90%).

HTS-metabarcoding procedures were followed as mentioned in Gaonkar et al., (2020). Only HTS metabarcodes haplotypes (OTUs) with reads more than 3 allocated to Chaetocerotaceae (Chaetoceros and Bacteriastrum) were used with a selection criterion of ≥ 90% similarity and ≥ 70% similarity after BLAST analysis using the protistan PR2 dataset. A total of 206 (n=82881) OTUs annotated as Chaetocerotacean haplotypes were obtained from the HRR dataset. See figures 1 and 2 in the Supplemental Files section.

BCO-DMO Processing Notes:
- added conventional header with dataset name, PI name, version date
- modified parameter names to conform with BCO-DMO naming conventions