An Illumina metabarcoding pipeline for fungi

Abstract

High-throughput metabarcoding studies on fungi and other eukaryotic microorganisms are rapidly becoming more frequent and more complex, requiring researchers to handle ever increasing amounts of raw sequence data. Here, we provide a flexible pipeline for pruning and analyzing fungal barcode (ITS rDNA) data generated as paired-end reads on Illumina MiSeq sequencers. The pipeline presented includes specific steps fine-tuned for ITS, that are mostly missing from pipelines developed for prokaryotes. It (1) employs state of the art programs and follows best practices in fungal high-throughput metabarcoding; (2) consists of modules and scripts easily modifiable by the user to ensure maximum flexibility with regard to specific needs of a project or future methodological developments; and (3) is straightforward to use, also in classroom settings. We provide detailed descriptions and revision techniques for each step, thus giving the user maximum control over data treatment and avoiding a black-box approach. Employing this pipeline will improve and speed up the tedious and error-prone process of cleaning fungal Illumina metabarcoding data.

Keywords: community ecology; data pruning; high throughput; internal transcribed spacer rDNA; next-generation sequencing.

Figure 1

Frequency of soil samples in relation to the number of Illumina MiSeq reads allocated to each sample. The majority of samples contained between 40,000 and 70,000 Illumina MiSeq reads.

Figure 2

Distribution of Illumina MiSeq reads and 97% OTUs across different taxonomic groups. Approximately, 5.9 million reads are assigned to fungi (from a total of >10 million reads). From a total of 14,636 OTUs 3208 could be assigned to fungi.