18S Amplicon sequence variants (ASVs) data of NEREA Augmented Observatory

The quality of the Illumina paired-end V9-18S raw reads (FASTQ format, 2 X 150 PE) was checked using vsearch (vsearch --fastq_stats), then pre-processed with cutadapt and vsearch to remove primer sequences, trim low quality bases and unify mixed orientation reads produced in the ligation-based library preparation. Processed reads were then used to generate amplicon sequence variants (ASVs) using the DADA2 R library; the pipeline was adapted from the one described on the program website (https://benjjneb.github.io/dada2/tutorial.html); no further quality filtering was implemented at this stage, except for discarding all reads with ambiguities (parameter maxN=0 of function filterAndTrim). Filtered F and R reads were used to train the error model and then denoised by applying the trained error model to generate ASVs. Finally, F and R reads were merged and checked for chimeras; up to 9 mismatches were allowed for read merging (parameter maxMismatch=9 of function mergePairs). ASVs were then classified with BLAST against the PR2 v5.01 reference database, integrated with 1293 sequences from Gulf of Naples protist strains and fungi environmental sequences. Highest bit score matches with the best taxonomic resolution were then selected among the returned results.