Earth Biogenome Project - Pilot Workflow

The primary workflow for the Earth Biogenome Project Pilot at NBIS.

Workflow overview

General aim:

flowchart LR
    hifi[/ HiFi reads /] --> data_inspection
    ont[/ ONT reads /] -->  data_inspection
    hic[/ Hi-C reads /] --> data_inspection
    data_inspection[[ Data inspection ]] --> preprocessing
    preprocessing[[ Preprocessing ]] --> assemble
    assemble[[ Assemble ]] --> validation
    validation[[ Assembly validation ]] --> curation
    curation[[ Assembly curation ]] --> validation

Current implementation:

flowchart TD
    input[/ Input file/] --> hifi
    input --> hic
    input --> taxonkit[[ TaxonKit name2taxid/reformat ]]
    taxonkit --> goat_taxon[[ GOAT taxon search ]]
    goat_taxon --> busco
    goat_taxon --> dtol[[ DToL lookup ]]
    hifi --> samtools_fa[[ Samtools fasta ]]
    samtools_fa --> fastk_hifi
    samtools_fa --> mash_screen
    hifi[/ HiFi reads /] --> fastk_hifi[[ FastK - HiFi ]]
    hifi --> meryl_hifi[[ Meryl - HiFi ]]
    hic[/ Hi-C reads /] --> fastk_hic[[ FastK - Hi-C ]]
    hifi --> meryl_hic[[ Meryl - Hi-C ]]
    assembly[/ Assembly /] --> quast[[ Quast ]]
    fastk_hifi --> histex[[ Histex ]]
    histex --> genescopefk[[ GeneScopeFK ]]
    fastk_hifi --> ploidyplot[[ PloidyPlot ]]
    fastk_hifi --> katgc[[ KatGC ]]
    fastk_hifi --> merquryfk[[ MerquryFK ]]
    assembly --> merquryfk
    meryl_hifi --> merqury[[ Merqury ]]
    assembly --> merqury
    fastk_hifi --> katcomp[[ KatComp ]]
    fastk_hic --> katcomp
    assembly --> busco[[ Busco ]]
    refseq_sketch[( RefSeq sketch )] --> mash_screen[[ Mash Screen ]]
    hifi --> mash_screen
    fastk_hifi --> hifiasm[[ HiFiasm ]]
    hifiasm --> assembly
    assembly --> purgedups[[ Purgedups ]]
    input --> mitoref[[ Mitohifi - Find reference ]]
    assembly --> mitohifi[[ Mitohifi ]]
    assembly --> fcsgx[[ FCS GX ]]
    fcs_fetchdb[( FCS fetchdb )] --> fcsgx
    mitoref --> mitohifi
    genescopefk --> quarto[[ Quarto ]]
    goat_taxon --> multiqc[[ MultiQC ]]
    quarto --> multiqc
    dtol --> multiqc
    katgc --> multiqc
    ploidyplot --> multiqc
    busco --> multiqc
    quast --> multiqc

Usage

nextflow run -params-file <params.yml> \
    [ -c <custom.config> ] \
    [ -profile <profile> ] \
    NBISweden/Earth-Biogenome-Project-pilot

where:

params.yml is a YAML formatted file containing workflow parameters such as input paths to the assembly specification, and settings for tools within the workflow.

Example:

input: 'assembly_spec.yml'
outdir: results
fastk: # Optional
  kmer_size: 31 # default 31
genescopefk: # Optional
  kmer_size: 31 # default 31
hifiasm: # Optional, default = no extra options: Key (e.g. 'opts01') is used in assembly build name (e.g., 'hifiasm-raw-opts01').
  opts01: "--opts A"
  opts02: "--opts B"
busco: # Optional, default: retrieved from GOAT
  lineages: 'auto' # comma separated string of lineages or auto.

Alternatively parameters can be provided on the command-line using the --parameter notation (e.g., --input <path> ).

<custom.config> is a Nextflow configuration file which provides additional configuration. This is used to customise settings other than workflow parameters, such as cpus, time, and command-line options to tools.

Example:

process {
    withName: 'BUSCO' {  // Selects the process to apply settings.
        cpus     = 6     // Overrides cpu settings defined in nextflow.config
        time     = 4.d   // Overrides time settings defined in nextflow.config to 4 days. Use .h for hours, .m for minutes.
        memory   = '20GB'  // Overrides memory settings defined in nextflow.config to 20 GB.
        // ext.args supplies command-line options to the process tool
        // overrides settings found in configs/modules.config
        ext.args = '--long'  // Supplies these as command-line options to Busco
    }
}

<profile> is one of the preconfigured execution profiles (uppmax, singularity_local, docker_local, etc: see nextflow.config). Alternatively, you can provide a custom configuration to configure this workflow to your execution environment. See Nextflow Configuration for more details.

Workflow parameter inputs

Mandatory:

input: A YAML formatted input file. Example assembly_spec.yml (See also test profile input TODO:: Update test profile):

sample:                          # Required: Meta data
  name: 'Laetiporus sulphureus'  # Required: Species name. Correct spelling is important to look up species information.
  ploidy: 2                      # Optional: Estimated ploidy (default: retrieved from GOAT)
  genome_size: 2345              # Optional: Estimated genome size (default: retrieved from GOAT)
  haploid_number: 13             # Optional: Estimated haploid chromosome count (default: retrieved from GOAT)
  taxid: 5630                    # Optional: Taxon ID (default: retrieved with Taxonkit)
  kingdom: Eukaryota             # Optional: (default: retrived with Taxonkit)
assembly:                        # Optional: List of assemblies to curate and validate.
  - assembler: hifiasm           # For each entry, the assembler,
    stage: raw                   # stage of assembly,
    id: uuid                     # unique id,
    pri_fasta: /path/to/primary_asm.fasta # and paths to sequences are required.
    alt_fasta: /path/to/alternate_asm.fasta
    pri_gfa: /path/to/primary_asm.gfa
    alt_gfa: /path/to/alternate_asm.gfa
  - assembler: ipa
    stage: raw
    id: uuid
    pri_fasta: /path/to/primary_asm.fasta
    alt_fasta: /path/to/alternate_asm.fasta
hic:                             # Optional: List of hi-c reads to QC and use for scaffolding
  - read1: '/path/to/raw/data/hic/LS_HIC_R001_1.fastq.gz'
    read2: '/path/to/raw/data/hic/LS_HIC_R001_2.fastq.gz'
hifi:                            # Required: List of hifi-reads to QC and use for assembly/validation
  - reads: '/path/to/raw/data/hifi/LS_HIFI_R001.bam'
rnaseq:                          # Optional: List of Rna-seq reads to use for validation
  - read1: '/path/to/raw/data/rnaseq/LS_RNASEQ_R001_1.fastq.gz'
    read2: '/path/to/raw/data/rnaseq/LS_RNASEQ_R001_2.fastq.gz'
isoseq:                          # Optional: List of Isoseq reads to use for validation
  - reads: '/path/to/raw/data/isoseq/LS_ISOSEQ_R001.bam'

Optional:

outdir: The publishing path for results (default: results).
publish_mode: (values: 'symlink' (default), 'copy') The file publishing method from the intermediate results folders (see Table of publish modes).
steps: The workflow steps to execute (default is all steps). Choose from:
- inspect: 01 - Read inspection
- preprocess: 02 - Read preprocessing
- assemble: 03 - Assembly
- purge: 04 - Duplicate purging
- polish: 05 - Error polishing
- screen: 06 - Contamination screening
- scaffold: 07 - Scaffolding
- curate: 08 - Rapid curation
- alignRNA: 09 - Align RNAseq data

Software specific:

Tool specific settings are provided by supplying values to specific keys or supplying an array of settings under a tool name. The input to -params-file would look like this:

input: assembly.yml
outdir: results
fastk:
  kmer_size: 31
genescopefk:
  kmer_size: 31
hifiasm:
  opts01: "--opts A"
  opts02: "--opts B"
busco:
  lineages: 'auto'

multiqc_config: Path to MultiQC configuration file (default: configs/multiqc_conf.yaml).

Uppmax and PDC cluster specific:

project: NAISS Compute allocation number.

Workflow outputs

All results are published to the path assigned to the workflow parameter results.

TODO:: List folder contents in results file

Customization for Uppmax

A custom profile named uppmax is available to run this workflow specifically on UPPMAX clusters. The process executor is slurm so jobs are submitted to the Slurm Queue Manager. All jobs submitted to slurm must have a project allocation. This is automatically added to the clusterOptions in the uppmax profile. All Uppmax clusters have node local disk space to do computations, and prevent heavy input/output over the network (which slows down the cluster for all). The path to this disk space is provided by the $SNIC_TMP variable, used by the process.scratch directive in the uppmax profile. Lastly the profile enables the use of Singularity so that all processes must be executed within Singularity containers. See nextflow.config for the profile specification.

The profile is enabled using the -profile parameter to nextflow:

nextflow run -profile uppmax <nextflow_script>

A NAISS compute allocation should also be supplied using the --project parameter.

Customization for PDC

A custom profile named dardel is available to run this workflow specifically on the PDC cluster Dardel. The process executor is slurm so jobs are submitted to the Slurm Queue Manager. All jobs submitted to slurm must have a project allocation. This is automatically added to the clusterOptions in the dardel profile. Calculations are performed in the scratch space allocated by PDC_TMP which is also on the lustre file system and is not node local storage. The path to this disk space is provided by the $PDC_TMP variable, used by the process.scratch directive in the dardel profile. Lastly the profile enables the use of Singularity so that all processes must be executed within Singularity containers. See nextflow.config for the profile specification.