Human de novo Assembly

Generate de novo assemblies from PacBio read data. Supports single-sample and trio-based assemblies.

Workflow for running de novo assembly using human PacBio whole genome sequencing (WGS) data. Written using Workflow Description Language (WDL). The assembly workflow performs de novo assembly on samples and trios.

Workflow Inputs

Each sample can independently have single-sample de novo assembly run. Additionally, if a trio is provided, trio-based assembly may be run.

Input Description

cohort

A cohort can include one or more samples. Samples need not be related.

cohort_id	A unique name for the cohort; used to name outputs
samples	The set of samples for the cohort. At least one sample must be defined.
run_de_novo_assembly_trio	Run trio binned de novo assembly.

samples

Sample information for each sample in the workflow run.

sample_id	A unique name for the sample; used to name outputs
movie_bams	The set of unaligned movie BAMs associated with this sample
sex	Sample sex
father_id	Paternal `sample_id`
mother_id	Maternal `sample_id`
run_de_novo_assembly	If true, run single-sample de novo assembly for this sample

reference

Files associated with the reference genome.

name	Reference name; used to name outputs (e.g., “GRCh38”)
fasta	Reference genome and index

backend

Backend where the workflow will be executed

zones

Zones where compute will take place; required if backend is set to ‘AWS’ or ‘GCP’.

aws_spot_queue_arn

Queue ARN for the spot batch queue; required if backend is set to ‘AWS’ and preemptible is set to true

aws_on_demand_queue_arn

Queue ARN for the on demand batch queue; required if backend is set to ‘AWS’ and preemptible is set to false

container_registry

Container registry where workflow images are hosted. If left blank, PacBio’s public Quay.io registry will be used.

preemptible

If set to true, run tasks preemptibly where possible. On-demand VMs will be used only for tasks that run for >24 hours if the backend is set to GCP. If set to false, on-demand VMs will be used for every task. Ignored if backend is set to HPC.

Workflow Outputs

The output set will depend on whether single-sample or trio-based de novo assembly is run.

Output Description

Sample de novo assembly

These files will be output if cohort.samples[sample] is set to true for any sample.

zipped_assembly_fastas	De novo dual assembly generated by hifiasm
assembly_noseq_gfas	Assembly graphs in GFA format.
assembly_lowQ_beds	Coordinates of low quality regions in BED format.
assembly_stats	Assembly size and NG50 stats generated by calN50.
asm_bam	minimap2 alignment of assembly to reference.
htsbox_vcf	Naive pileup variant calling of assembly against reference with `htsbox`
htsbox_vcf_stats	`bcftools stats` summary statistics for `htsbox` variant calls

Trio de novo assembly

These files will be output if cohort.de_novo_assembly_trio is set to true and there is at least one parent-parent-kid trio in the cohort.

trio_zipped_assembly_fastas	Haplotype-resolved de novo assembly of the trio kid generated by hifiasm with trio binning
trio_assembly_noseq_gfas	Assembly graphs in GFA format.
trio_assembly_lowQ_beds	Coordinates of low quality regions in BED format.
trio_assembly_stats	Assembly size and NG50 stats generated by calN50.
trio_asm_bams	minimap2 alignment of assembly to reference.
haplotype_key	Indication of which haplotype (`hap1`/`hap2`) corresponds to which parent.

References

Reference datasets are hosted publicly for use in the pipeline.

Containers

Docker images definitions used by this workflow can be found in the wdl-dockerfiles repository. Images are hosted in PacBio’s quay.io. Docker images used in the workflow are pegged to specific versions by referring to their digests rather than tags.