2). Overview

Table of contents

• Simplified schematic overview
• Description

Simplified schematic overview

Description

The pipeline will perform the following steps:

1). Read quality control

• Assess pre-QC read quality using NanoPlot and Seqtk.
• Remove adaptors from reads using Porechop.
• Filter long reads by quality using Filtlong.

2). Read-based taxonomic annotation

• Read-based taxonomic classification using Kraken2.
• Read-based taxonomic classification using Centrifuger.
• Bayesian Reestimation of Abundance of Kraken2 and Centrifuger outputs using Bracken.
• Standardise taxonomic profiles output by Kraken2, Centrifuger and Bracken using taxpasta.
• Read-based taxonomic classification using Sylph.
• Merge taxonomic classification results into summary reports.

3). Host read removal

• Identify host-contaminant reads using Kraken2. and/or Minimap2 against a host reference or genome, respectively.
• Remove host-contaminant reads using SeqKit.
• Assess post-QC read quality using NanoPlot and Seqtk.
• Merge pre- and post-read QC metrics into a summary report.

4). Assembly

• Assembly post-QC host-decontaminated reads using Flye.
• Map reads to the full metagenome assembly using Minimap2 and polish (contig error correction) with Racon (0-4 rounds of correction).
• Polish full metagenome assembly using Medaka.

5). Contig analysis

• Identify viral, proviral and plasmid sequences in across all contigs using geNomad.
• Identify closest taxonomic hits to each contig using skani.
• Calculate contig length and GC content using SeqKit.
• Calculate read depth (coverage) using Samtools.

6). Assembly

• Classify contigs into classes: archaea, bacteria, prokarya, eukarya, organelle (mitochondria, plastid) or unknown using Tiara.
• Perform metagenomic binning, separate out contigs into individual metagenome assembled genomes (MAGs), using both SemiBin2 and MetaDecoder.
• Generate consensus bins from SemiBin2 and MetaDecoder and outputs using DAS Tool

7). Bin quality control

• Assess the quality of genome bins using CheckM.
• Assign taxonomic classifications to each MAG using GTDB-Tk.
• Assess the quality of genome bins using CheckM.
• Calculate assembly statistics using assembly_stats.py.
• Merge bin QC and contig QC results into summary reports.

8). Typing

Subset MAGs of interest (target species) using filter_gtdbtk.py and dependent on taxonomic classification pass them on to individual subworkflows (run per-MAG).

8a). Bacteria

• Identify virulence genes using VirulenceFinder.
• Assign sequence type to MAGs using MLST.
• Assign sequence type to MAG specific readsets using Krocus.
• Screen for genes of interest using BLASTN.
• Screen for genes of interest using Genefinder.
• Identify plasmids using PlasmidFinder.

8b). Listeria monocytogenes

• Perform in silico serogroup typing prediction using LisSero.

8c). Salmonella

• Perform in silico Salmonella serotyping using SeqSero2.
• Predict serovar, antigen gene and cgMLST alleles using SISTR.
• Identify antimicrobial resistance genes and lineages of S. Typhi and S. Paratyphi B using Mykrobe.

8d). Escherichia coli / Shigella spp.

• Differentiate Shigella/Enteroinvasive Escherichia coli and identify serotype using ShigEIFinder.
• Determine Shigella serotype using ShigaTyper.
• Determine E. coli serotype using ECTyper.
• Identify serotype of Shigatoxin producing E. coli using STECFinder.
• Identify antimicrobial resistance genes and lineages of S. sonnei using Mykrobe.

9). Antimicrobial resistance

• Identify AMR genes (incl. point mutations) and virulence/stress resistance genes using AMRFinderPlus.
• Identify AMR and virulence resistance genes across multiple databases using ABRicate.
• Identify AMR genes (incl.point mutations) using RGI.
• Identify AMR genes using ResFinder.
• Identify AMR conferring point mutations using PointFinder.
• Merge the AMR results into a summary report. Finally, all major results are merged into a single summary report, which include the per-bin typing information produce in step 8).