Swift’s ITS analysis workflow using Qiime 2

Software package requirements: Qiime 2 (https://qiime2.org/)

Setup:

1. Clone the package using "git clone https://github.com/swiftbiosciences/q2_ITS.git" to get following files: q2wkflow_ITS_ver3.sh, mergeStrand_taxonomy.py, config_SNAP_ITS.txt, primer file, and reference file (1) for the naive baysian classifier using UNITE ITS reference version 8.

2. Edit “config_SNAP_ITS.txt” to enter correct absolute paths to each tool, primer file, and expected read length after primer is trimmed. Set "AS_PE" to 'true' if you wish to merge the paired end reads, i.e. R1 and R2, and process the merged reads (recommended for longer read lengths, e.g. 2X250, 2X300).

   Command to run: q2wkflow_ITS_ver3.sh config_SNAP_ITS.txt inputdir workdir

3. Results organization: ASV feature table can be found in "Export" directory; "QObj" directory contains intermediate files and tables some of which could be imported into Qiime2 web interface for further exploration and visualization; "Fastq" directory contains intermediate fastq files at all different stages.

Additional notes:

1. This workflow is intended to serve as a basic example using Qiime 2 for processing fungal ITS data prepared using Swift SNAP ITS1 library kit. Users should feel free to make any modifications for improvements.

2. Additional quality trimming/filtering of reads may be needed depending on specific sequencing data sets.

3. The performance of this workflow is highly dependent on the taxonomy classifier chosen (this example uses naive baysian classifier with UNITE ITS reference set version 8 included in the package, for ITS analysis). Users should feel free to obtain and test other type of classifiers & reference sets that best fit the environments sampled.