Routine Assembly

A generic pipeline for creating routine draft assemblies

Analyses

• Read trimming & QC: fastp and filtlong
• Genome Assembly: shovill, unicycler or dragonflye
• Gene Annotation: prokka or bakta
• Assembly QC: quast

Usage

By default, shovill and prokka will be used:

nextflow run BCCDC-PHL/routine-assembly-nf \
  --fastq_input <fastq input directory> \
  --outdir <output directory>

Unicycler and/or bakta can be used with the --unicycler and --bakta flags:

nextflow run BCCDC-PHL/routine-assembly-nf \
  --fastq_input <fastq input directory> \
  --unicycler \
  --bakta \
  --outdir <output directory>

Any combination of shovill/unicycler and prokka/bakta is supported: Shovill with bakta:

nextflow run BCCDC-PHL/routine-assembly-nf \
  --fastq_input <fastq input directory> \
  --bakta \
  --outdir <output directory>

Unicycler with prokka:

nextflow run BCCDC-PHL/routine-assembly-nf \
  --fastq_input <fastq input directory> \
  --unicycler \
  --outdir <output directory>

The pipeline also supports a 'samplesheet input' mode. Pass a samplesheet.csv file with the headers ID, R1, R2:

nextflow run BCCDC-PHL/routine-assembly-nf \
  --samplesheet_input <samplesheet.csv> \
  --outdir <output directory>

Eg:

ID,R1,R2
sample-01,/path/to/sample-01_R1.fastq.gz,/path/to/sample-01_R2.fastq.gz
sample-02,/path/to/sample-02_R1.fastq.gz,/path/to/sample-02_R2.fastq.gz
sample-03,/path/to/sample-03_R1.fastq.gz,/path/to/sample-03_R2.fastq.gz

Hybrid Assembly Mode

If long (Oxford Nanopore) and short (illumina) reads are both available, hybrid assemblies can be performed with unicycler. Note that shovill does not support hybrid assemblies. Add the --hybrid flag to perform hybrid assembly, and supply long reads using the --long_reads flag. Sample IDs for both long and short reads are taken from all characters of the fastq files up to the first underscore _. Sample IDs for short and long reads must match in order for the pipeline to match them up for hybrid assembly.

nextflow run BCCDC-PHL/routine-assembly-nf \
  --fastq_input <fastq input directory> \
  --long_reads <long reads input directory> \
  --unicycler \
  --hybrid \
  --outdir <output directory>

Hybrid assembly mode is compatible with samplesheet input mode. When using a samplesheet for hybrid assemblies, an additional field with header LONG is required. Note that in this mode, because the sample IDs are explicitly provided in the samplesheet, it isn't strictly necessary that the short and long read filenames have matching sample IDs (though it's still probably good practice to do so).

Eg:

ID,R1,R2,LONG
sample-01,/path/to/sample-01_R1.fastq.gz,/path/to/sample-01_R2.fastq.gz,/path/to/sample-01_L.fastq.gz
sample-02,/path/to/sample-02_R1.fastq.gz,/path/to/sample-02_R2.fastq.gz,/path/to/sample-02_L.fastq.gz
sample-03,/path/to/sample-03_R1.fastq.gz,/path/to/sample-03_R2.fastq.gz,/path/to/sample-03_L.fastq.gz

All samples in the samplesheet should have both short and long reads when running in hybrid assembly mode.

Run the pipeline as follows:

nextflow run BCCDC-PHL/routine-assembly-nf \
  --samplesheet /path/to/samplesheet.csv \
  --unicycler \
  --hybrid \
  --outdir <output directory>

Long-read-only Assembly Mode

If only long (Oxford Nanopore) reads are available, a long-read-only assembly mode is supported using the dragonflye assembler.

nextflow run BCCDC-PHL/routine-assembly-nf \
  --long_reads <long reads input directory> \
  --dragonflye \
  --outdir <output directory>

Output

An output directory will be created for each sample under the directory provided with the --outdir flag. The directory will be named by sample ID, inferred from the fastq files (all characters before the first underscore in the fastq filenames).

If we have sample-01_R{1,2}.fastq.gz, the output directory will be:

sample-01
├── sample-01_20211125165316_provenance.yml
├── sample-01_fastp.csv
├── sample-01_fastp.json
├── sample-01_shovill_prokka.gbk
├── sample-01_shovill_prokka.gff
├── sample-01_shovill_quast.csv
├── sample-01_shovill.fa
└── sample-01_shovill.log

Including the tool name suffixes to output files allows re-analysis of the same sample with multiple tools without conflicting output filenames:

sample-01
├── sample-01_20211125165316_provenance.yml
├── sample-01_20211128122118_provenance.yml
├── sample-01_unicycler_bakta.gbk
├── sample-01_unicycler_bakta.gff
├── sample-01_unicycler_bakta.json
├── sample-01_unicycler_bakta.log
├── sample-01_fastp.csv
├── sample-01_fastp.json
├── sample-01_shovill_prokka.gbk
├── sample-01_shovill_prokka.gff
├── sample-01_shovill_quast.csv
├── sample-01_unicycler_quast.csv
├── sample-01_shovill.fa
├── sample-01_shovill.log
├── sample-01_unicycler.fa
├── sample-01_unicycler.gfa
└── sample-01_unicycler.log

If the --versioned_outdir flag is used, then a sub-directory will be created below each sample, named with the pipeline name and minor version:

sample-01
    └── routine-assembly-v0.2-output
        ├── sample-01_20220216172238_provenance.yml
        ├── sample-01_fastp.csv
        ├── sample-01_fastp.json
        ├── sample-01_shovill.fa
        ├── sample-01_shovill.log
        ├── sample-01_shovill_prokka.gbk
        ├── sample-01_shovill_prokka.gff
        └── sample-01_shovill_quast.csv

This is provided as a way of combining outputs of several different pipelines or re-analysis with future versions of this pipeline:

sample-01
    └── routine-assembly-v0.2-output
    │   ├── sample-01_20220216172238_provenance.yml
    │   ├── sample-01_fastp.csv
    │   ├── sample-01_fastp.json
    │   ├── sample-01_shovill.fa
    │   ├── sample-01_shovill.log
    │   ├── sample-01_shovill_prokka.gbk
    │   ├── sample-01_shovill_prokka.gff
    │   └── sample-01_shovill_quast.csv
    └── routine-assembly-v0.3-output
        ├── sample-01_20220612091224_provenance.yml
        ├── sample-01_fastp.csv
        ├── sample-01_fastp.json
        ├── sample-01_shovill.fa
        ├── sample-01_shovill.log
        ├── sample-01_shovill_prokka.gbk
        ├── sample-01_shovill_prokka.gff
        └── sample-01_shovill_quast.csv

Provenance files

For each pipeline invocation, each sample will produce a provenance.yml file with the following contents:

- pipeline_name: BCCDC-PHL/routine-assembly
  pipeline_version: 0.3.0
- timestamp_analysis_start: 2021-11-25T16:53:10.549863
- input_filename: sample-01_R1.fastq.gz
  sha256: 4ac3055ac5f03114a005aff033e7018ea98486cbebdae669880e3f0511ed21bb
  file_type: fastq-input
- input_filename: sample-01_R2.fastq.gz
  sha256: 8db388f56a51920752319c67b5308c7e99f2a566ca83311037a425f8d6bb1ecc
  file_type: fastq-input
- process_name: fastp
  tools:
    - tool_name: fastp
      tool_version: 0.23.1
- process_name: shovill
  tools:
    - tool_name: shovill
      tool_version: 1.1.0
- process_name: prokka
  tools:
    - tool_name: prokka
      tool_version: 1.14.5
      parameters:
        - parameter: --compliant
          value: null
- process_name: quast
  tools:
    - tool_name: quast
      tool_version: 5.0.2
      parameters:
        - parameter: --space-efficient
          value: null
        - parameter: --fast
          value: null

The filename of the provenance file includes a timestamp with format YYYYMMDDHHMMSS to ensure that re-analysis of the same sample will create a unique provenance.yml file.