Long Reads Analysis Mode

Long Reads Analysis Mode#

The long reads analysis mode (--type longreads) is designed for processing PacBio or Nanopore long-read sequencing data. This mode uses Flye for metagenomic assembly, which is optimized for long-read technologies.

Overview#

The long reads mode is optimized for third-generation sequencing technologies (PacBio and Nanopore) that produce longer reads compared to Illumina short-read sequencing. Longer reads enable better assembly of complex metagenomic communities and improved detection of complete gene clusters.

Workflow#

The long reads workflow consists of the following main steps:

  1. Quality Control (FastQC, optional for DNA) - FastQC quality assessment (primarily for RNA-seq data) - DNA long reads typically skip QC/trimming steps as they are less affected by adapters

  2. Taxonomic Filtering (Kraken2, optional) - Taxonomic classification using Kraken2 (applied to RNA-seq data) - Extraction of reads matching specified taxonomy - Can be skipped with --skip_kraken_extraction

  3. Assembly (Flye) - Long-read metagenomic assembly using Flye - Supports multiple Flye modes for different sequencing technologies - Configurable via --flye_mode parameter

  4. Gene Prediction (Pyrodigal) - Prodigal-based gene finding optimized for metagenomic data - Generates protein sequences (FAA) and gene annotations (GFF)

  5. CAZyme Annotation (run_dbCAN) - CAZyme identification using dbCAN database - CGC (CAZyme Gene Cluster) detection - Substrate prediction

  6. Read Mapping (Minimap2 for DNA, BWA-MEM for RNA) - Mapping of long DNA reads back to assembled contigs using Minimap2 - Mapping of RNA reads (if provided) using BWA-MEM for expression analysis - Coverage calculation for genes and CGCs

  7. Abundance Calculation - Gene-level abundance calculation based on read coverage - CGC abundance and visualization - Generation of bar plots and heatmaps

  8. Report Generation (MultiQC) - Aggregated quality control and analysis reports

Usage#

Note

Before running the pipeline, make sure you have cloned the repository. See Nextflow Pipeline: Usage for installation instructions.

Basic Usage#

The simplest command to run long reads analysis (assuming you are in the dbcan-nf directory):

nextflow run main.nf \
  --input samplesheet.csv \
  --outdir results \
  --type longreads \
  -profile docker \
  --skip_kraken_extraction # based on the database size of kraken2, you can skip this step if the database is too large.

Flye Mode Selection#

The --flye_mode parameter allows you to specify the appropriate Flye mode for your sequencing technology:

# PacBio HiFi reads (default)
nextflow run main.nf \
  --input samplesheet.csv \
  --outdir results \
  --type longreads \
  --flye_mode --pacbio-hifi \
  -profile docker

# PacBio raw reads
nextflow run main.nf \
  --input samplesheet.csv \
  --outdir results \
  --type longreads \
  --flye_mode --pacbio-raw \
  -profile docker

# Nanopore raw reads
nextflow run main.nf \
  --input samplesheet.csv \
  --outdir results \
  --type longreads \
  --flye_mode --nano-raw \
  -profile docker

# Nanopore high-quality reads
nextflow run main.nf \
  --input samplesheet.csv \
  --outdir results \
  --type longreads \
  --flye_mode --nano-hq \
  -profile docker

With RNA-seq Data#

When RNA-seq transcriptome data is provided in the samplesheet, the pipeline will:

  • Process RNA reads through quality control (FastQC + TrimGalore)

  • Apply Kraken2 taxonomic filtering (if enabled)

  • Map RNA reads to assembled contigs using BWA-MEM

  • Calculate RNA-based abundance for expression analysis

  • Generate separate DNA and RNA abundance plots

Example with RNA-seq:

nextflow run main.nf \
  --input samplesheet_with_rna.csv \
  --outdir results \
  --type longreads \
  --flye_mode --pacbio-hifi \
  -profile docker

Skipping Steps#

You can skip certain steps if needed:

nextflow run main.nf \
  --input samplesheet.csv \
  --outdir results \
  --type longreads \
  --skip_kraken_extraction \
  -profile docker

Flye Mode Options#

The --flye_mode parameter accepts the following Flye assembly modes:

Flye Mode Options#

Mode

Description

--pacbio-hifi

PacBio HiFi (high-fidelity) reads. Default mode. Best for high-quality PacBio data.

--pacbio-raw

PacBio raw (CLR) reads. Use for standard PacBio sequencing data.

--nano-raw

Nanopore raw reads. Use for standard Nanopore sequencing data.

--nano-hq

Nanopore high-quality reads. Use for Q20+ or similar high-quality Nanopore data.

Output Files#

The pipeline generates output files organized in the following directory structure:

Assembly Results#

  • flye/ - *_assembly.fasta.gz: Assembled contigs in FASTA format (gzipped) - Assembly statistics and reports

Gene Prediction#

  • pyrodigal/ - *.faa.gz: Predicted protein sequences (gzipped) - *.gff.gz: Gene annotations in GFF format (gzipped)

CAZyme Annotation#

  • rundbcan/ - *_dbcan/: Directory containing all run_dbCAN results

    Files in this directory:

    • overview.tsv: CAZyme annotation overview

    • dbCAN_hmm_results.tsv: HMM-based CAZyme predictions

    • dbCANsub_hmm_results.tsv: Subfamily predictions

    • diamond.out: DIAMOND search results

    • *_cgc.gff: CGC annotations

    • *_cgc_standard_out.tsv: CGC standard output

    • *_substrate_prediction.tsv: Substrate predictions

    • *_synteny_pdf/: Synteny plots for CGCs

Read Mapping#

  • minimap2/: Minimap2 alignment files for long DNA reads - *.bam: Aligned reads - *.bam.bai: BAM index files

  • bwa_index_mem/: BWA-MEM alignment files for RNA reads (if provided) - *.bam: Aligned reads - *.bam.bai: BAM index files

Coverage and Abundance#

  • dbcan_utils_cal_coverage/ - *_depth.txt: Gene coverage depth files

  • dbcan_utils_cal_abund/ - *_abund/: Abundance calculation results

    Files in this directory:

    • *_abund.txt: Gene abundance values

    • *_cgc_abund.txt: CGC abundance values

  • dbcan_plot/ - *_pdf/: Visualization plots

    Files in this directory:

    • heatmap.pdf: Abundance heatmap

    • ec.pdf: EC number distribution

    • family.pdf: CAZyme family distribution

    • subfamily.pdf: Subfamily distribution

  • cgc_depth_plot/ - *_cgc_depth.tsv: CGC depth coverage data - *_cgc_depth.pdf: CGC depth plots

Quality Control#

  • fastqc/: FastQC reports (for RNA-seq data)

  • trimgalore/: TrimGalore reports (for RNA-seq data)

  • multiqc/: MultiQC aggregated report

Pipeline Information#

  • pipeline_info/: Execution reports, parameters, and software versions

Key Features#

  • Long-Read Optimized: Uses Flye assembler optimized for PacBio and Nanopore reads

  • Flexible Flye Modes: Supports multiple Flye modes for different sequencing technologies

  • Dual Mapping: Minimap2 for long DNA reads, BWA-MEM for RNA reads

  • Complete Gene Clusters: Longer reads enable better assembly of complete CGCs

  • RNA-seq Integration: Optional RNA-seq support for expression analysis

Advantages of Long Reads#

  • Better Contiguity: Longer contigs due to long-read sequencing

  • Complete Genes: More complete gene predictions, especially for large genes

  • Reduced Fragmentation: Fewer fragmented gene clusters

  • Repeat Resolution: Better resolution of repetitive regions

  • Structural Variants: Improved detection of structural variations

Considerations#

  • Computational Requirements: Long-read assembly typically requires more memory and time

  • Error Rates: Long reads may have higher error rates than short reads

  • Coverage: Lower coverage requirements compared to short reads

  • Cost: Long-read sequencing may be more expensive per base

Best Practices#

  1. Choose Correct Mode: Select the appropriate --flye_mode for your sequencing technology

  2. Quality Assessment: Review FastQC reports for RNA-seq data

  3. Coverage Planning: Long reads require less coverage than short reads (typically 20-30x)

  4. Resource Allocation: Ensure sufficient memory for Flye assembly

When to Use Long Reads Mode#

Recommended: - PacBio or Nanopore sequencing data - When complete gene clusters are important - When dealing with complex or repetitive regions - When structural variation detection is needed

Not Recommended: - Illumina short-read data (use --type shortreads instead) - When computational resources are very limited - When only basic CAZyme annotation is needed

Example Results#

For example visualizations from long reads mode, see Nextflow Pipeline: Results Examples.

See Also#

Contents