metashot/kraken2 is a workflow for the taxonomic classification of reads and the abundance estimation of species in metagenomic samples.

• MetaShot Home

• Input: single-end, paired-end (also interleaved) Illumina sequences (gzip and bzip2 compressed FASTA or FASTQ also supported);
• Histogram text files (for each input sample) of base frequency, quality scores, GC content, average quality and length are generated from input reads using bbduk;
• Taxonomic classification using Kraken 2;
• Abundance estimation for each taxonomic level using Bracken.

1. Install Docker (or Singulariry) and Nextflow (see Dependences);
2. Download and extract/unzip a Kraken 2 / Bracken database available at https://benlangmead.github.io/aws-indexes/k2;
3. Start running the analysis:

nextflow run metashot/kraken2 \
  --reads '*_R{1,2}.fastq.gz' \
  --kraken2_db k2db \
  --read_len 100 \
  --outdir results

See the file nextflow.config for the complete list of parameters.

The files and directories listed below will be created in the results directory after the pipeline has finished.

• combined.kraken2.report: combined kraken2 report;
• combined.kraken2.mpa: combined kraken2 report in mpa (MetaPhlAn) format;
• combined_bracken: contains the combined bracken outputs (one file for each taxonomic level, from Domain D to Species S);
• combine_bracken_reports: same as combined_bracken but in kraken2 report format;
• combine_bracken_mpa: same as combined_bracken but in mpa (MetaPhlAn) format;

• raw_reads_stats: base frequency, quality scores, gc content, average quality and length for each input sample;
• kraken2: kraken2 output for each sample.
• bracken: bracken output for each taxonomic level (from Domain D to Species S) for each sample.

Kraken 2 requires enough free memory to hold the index in RAM. If the index size is 47 GB (standard database, 2020/09/19) you will need slightly more than that free in RAM (set the --max_memory parameter to 64.GB).

Please refer to System requirements for the complete list of system requirements options.