hifi-solves-run-humanwgs 2.8.3

Upload required files and run [PacBio's Human WGS workflow](https://github.com/PacificBiosciences/HiFi-human-WGS-WDL) via [DNAstack's Workbench](https://omics.ai/workbench/)

HiFi Solves Human WGS workflow runner

As part of the HiFi Solves Consortium, organizations will run their sequencing data through PacBio's Human Whole Genome Sequencing (WGS) pipeline.

This package handles uploading all required raw data to the organization's cloud, configures the required workflow metadata, and triggering a run of the HumanWGS workflow. Output files are automatically ingested into Publisher and made available on hifisolves.org.

Package information

• HumanWGS pipeline version: v2.1.1

Requirements

• python3.13+
• An engine registered on Workbench
• Credentials for the relevant backend (supported backends: AWS, Azure, GCP)

Installation

python3 -m pip install hifi-solves-run-humanwgs

Script usage

Arguments

usage: hifisolves-ingest [-h] [-v] [-s SAMPLE_INFO] [-m MOVIE_BAMS] [-c FAM_INFO] -b {AWS,GCP,AZURE} -r REGION -o ORGANIZATION [-e ENGINE] [-u] [-i] [-f]
                         [--families FAMILIES] [--aws-storage-capacity AWS_STORAGE_CAPACITY]

Upload genomics data and run PacBio's official Human WGS pipeline

options:
  -h, --help            show this help message and exit
  -v, --version         Program version
  -b {AWS,GCP,AZURE}, --backend {AWS,GCP,AZURE}
                        Backend where infrastructure is set up
  -r REGION, --region REGION
                        Region where infrastructure is set up
  -o ORGANIZATION, --organization ORGANIZATION
                        Organization identifier; used to infer bucket names
  -e ENGINE, --engine ENGINE
                        Engine to use to run the workflow. Defaults to the default engine set in Workbench.
  -u, --upload-only     Upload movie BAMs and generate inputs JSON only; do not submit the workflow. If set, --write-inputs-json will also be set automatically.
  -i, --write-inputs-json
                        Write inputs JSON and engine configuration to a file. Files will be named {family_id}.inputs.json, {family_id}.engine_params.json, {family_id}.run_tags.json.
  -f, --force-rerun-failed
                        Force rerun samples that have previously been run and failed; will not rerun samples that are currently running or have a succeeded run.
  --families FAMILIES   Comma-separated list of family IDs to process. If not specified, all families in the sample info file will be processed.
  --aws-storage-capacity AWS_STORAGE_CAPACITY
                        Storage capacity override for AWS HealthOmics backend. Defaults to total size of input BAMs across all samples * 8. Supply either the requested storage capacity in GB, or 'DYNAMIC' to set storage to dynamic.

Sample information:
  Provide either --sample-info, OR both --movie-bams and --fam-info

  -s SAMPLE_INFO, --sample-info SAMPLE_INFO
                        Path to sample info CSV or TSV. This file should have columns [family_id, sample_id, movie_bams, father_id, mother_id, sex]. See documentation for more information on the format of this file.
  -m MOVIE_BAMS, --movie-bams MOVIE_BAMS
                        Path to movie bams CSV or TSV. This file should have columns [sample_id, movie_bams]. Repeated rows for each sample can be added if the sample has more than one associated movie bam.
  -c FAM_INFO, --fam-info FAM_INFO
                        Path to family information. This file should have columns [family_id, sample_id, father_id, mother_id, sex]. It can optionally have additional phenotype columns (columns 6-end), but this information will not be used.

Sample info file

The sample info file defines the set of samples that will be run through the workflow. The workflow can either be run on individual samples or on families (typically trios, where sequencing data exists for the mother, father, and child). One workflow run will be submitted for each unique family ID, including all samples that share that family ID.

This information is organized into a CSV file with the following columns:

Column name	Description
family_id	Unique identifier for this family / cohort. If you are running a single sample through the workflow, this can be set to the same value as sample_id.
sample_id	Sample identifier
movie_bams†	Local path to a BAM file (either movie BAM or aligned BAM) associated with this sample
father_id	sample_id of the father. This field can be left blank if the sample's father is not included in the run.
mother_id	sample_id of the mother. This field can be left blank if the sample's mother is not included in the run.
sex	Set to either "MALE" or "FEMALE"

† There can be more than one BAM for a sample. If this is the case, a new row should be generated for each additional movie_bam; family_id and sample_id must be set for these fields, but information from other fields need not be repeated.

Example sample info files

All samples for all runs (singleton and family-based runs) may be included in a single sample info file; a separate run will be submitted for every unique family ID in the sample_info CSV.

Singleton

Here we have a single sample, HG005, with two associated movie bams found at the local paths bams/HG005_1.hifi_reads.bam and bams/HG005_2.hifi_reads.bam. The sample is being run alone so father_id and mother_id are left blank. Sex information only needs to be included once and can be omitted for further rows associated with the same sample_id.

family_id,sample_id,movie_bams,father_id,mother_id,sex
HG005,HG005,bams/HG005_1.hifi_reads.bam,,,MALE
HG005,HG005,bams/HG005_2.hifi_reads.bam,,,

Trio

Here we have a trio of samples: a child (HG005), father (HG006), and mother (HG007). The mother and father samples have several associated movie_bams, so there are multiple rows for each.

family_id,sample_id,movie_bams,father_id,mother_id,sex
hg005-trio,HG005,bams/HG005_1.hifi_reads.bam,HG006,HG007,MALE
hg005-trio,HG006,bams/HG006_1.hifi_reads.bam,,,MALE
hg005-trio,HG006,bams/HG006_2.hifi_reads.bam,,,
hg005-trio,HG007,bams/HG007_1.hifi_reads.bam,,,FEMALE
hg005-trio,HG007,bams/HG007_2.hifi_reads.bam,,,
hg005-trio,HG007,bams/HG007_3.hifi_reads.bam,,,

Alternative to the sample info file - --movie-bams and --fam-info

Instead of providing a --sample-info file, you may choose to organize your information into two separate files: --movie-bams, and --fam-info.

Movie bams

Provided using the --movie-bams argument.

Column name	Description
sample_id	Sample identifier
movie_bams†	Local path to a movie BAM file associated with this sample

† There can be more than one movie bam for a sample. If this is the case, a new row should be generated for each additional movie_bam.

Example movie bam file

sample_id,movie_bams
HG005,bams/HG005_1.hifi_reads.bam
HG006,bams/HG006_1.hifi_reads.bam
HG006,bams/HG006_2.hifi_reads.bam
HG007,bams/HG007_1.hifi_reads.bam
HG007,bams/HG007_2.hifi_reads.bam
HG007,bams/HG007_3.hifi_reads.bam

Family information

Provided using the --fam-info argument. This file is related to PLINK's fam info format, with some modifications (namely, a header is required, and there can be multiple (or zero) phenotypes columns; note that phenotype information is discarded here).

Column name	Description
family_id	Unique identifier for this family / cohort. If you are running a single sample through the workflow, this can be set to the same value as sample_id.
sample_id	Sample identifier
father_id	sample_id of the father. This field can be left blank if the sample's father is not included in the run.
mother_id	sample_id of the mother. This field can be left blank if the sample's mother is not included in the run.
sex	1=male, 2=female

Example fam info file

family_id,sample_id,father_id,mother_id,sex,HP:0001250,HP:0001263
hg005-trio,HG005,HG006,HG007,1,2,2
hg005-trio,HG006,,,1,1,1
hg005-trio,HG007,,,2,1,1

Running the script

By default, the script will both upload input files and trigger workflow runs.

To upload input files only, run using the --upload-only flag.

Environment Variables

WORKBENCH_URL

By default, the script will use workbench.omics.ai as the Workbench URL. You can override this by setting the WORKBENCH_URL environment variable:

export WORKBENCH_URL="custom-workbench.example.com"

Filtering to specific families

Use the --families option to process only a subset of families from your sample info file. This is useful for:

• Rerunning specific families that failed with adjusted parameters (e.g., memory overrides)
• Testing the workflow with a subset of families before processing all data
• Uploading data for specific families only

# Process a single family
hifisolves-ingest \
    --sample-info sample_info.csv \
    --families "hg005-trio" \
    --backend aws \
    --region "${AWS_REGION}" \
    --organization "${ORGANIZATION}"

# Process multiple families
hifisolves-ingest \
    --sample-info sample_info.csv \
    --families "hg005-trio,hg006-trio,singleton-1" \
    --backend aws \
    --region "${AWS_REGION}" \
    --organization "${ORGANIZATION}"

# Rerun failed families with increased memory
hifisolves-ingest \
    --sample-info sample_info.csv \
    --families "hg005-trio" \
    --force-rerun-failed \
    --hiphase-override-mem-gb 128 \
    --backend aws \
    --region "${AWS_REGION}" \
    --organization "${ORGANIZATION}"

If any specified family is not found in the sample info file, the script will exit with an error listing the missing families and available families.

Example run command - AWS

The AWS HealthOmics backend requires storage capacity for the run to be set; this capacity includes all inputs, outputs, and intermediate workflow files that are generated during workflow execution. The script will attempt to estimate the required capacity based on the size of input files, but this value can be overridden by setting the --aws-storage-capacity flag to either:

• 'DYNAMIC': storage will scale dynamically as the workflow runs; you should theoretically not run out of storage space
• <storage_capacity_gb>: the storage capacity in GB, between 0 and 9600 (9.6 TB)

See the HealthOmics docs for more information on run storage.

# AWS credentials
export AWS_ACCESS_KEY_ID=""
export AWS_SECRET_ACCESS_KEY=""
export AWS_SESSION_TOKEN=""

AWS_REGION=""
# Used for naming upload and output buckets
ORGANIZATION=""

hifisolves-ingest \
    --sample-info sample_info.csv \
    --backend aws \
    --region "${AWS_REGION}" \
    --organization "${ORGANIZATION}"

Example run command - Azure

# Azure credentials; needs Read, Add, Write, Create, Delete, List
export AZURE_STORAGE_SAS_TOKEN=""

AZURE_REGION=""
# Used for naming upload and output buckets; this is going to be == the storage account name
ORGANIZATION=""

hifisolves-ingest \
    --sample-info sample_info.csv \
    --backend Azure \
    --region "${AZURE_REGION}" \
    --organization "${ORGANIZATION}"

If you have files already uploaded in the target storage account, their paths may be referenced in the format /<storage_account>/rawdata/path/to/file.

Copying Azure <> Azure

If source files are currently in cloud storage, they can be copied into the target storage account rather than copying from local -> cloud.

BAM URLs in the sample_info CSV file should be in the format /<src_storage_account>/<src_storage_container>/path/to/movie.bam.

An additional env variable, SOURCE_CONTAINER_SAS_TOKEN, should be defined. This SAS token should have Read and List permissions on the source container.

# SAS token for the source bucket (R/L)
export SOURCE_CONTAINER_SAS_TOKEN=""

# SAS token for the destination bucket (R/A/W/C/D/L)
export AZURE_STORAGE_SAS_TOKEN=""

AZURE_REGION=""
ORGANIZATION=""

hifisolves-ingest \
    --sample-info sample_info.csv \
    --backend Azure \
    --region "${AZURE_REGION}" \
    --organization "${ORGANIZATION}"

Example run command - GCP

# GCP credentials - GOOGLE_APPLICATION_CREDENTIALS should point towards a JSON file containing service account information
export GOOGLE_APPLICATION_CREDENTIALS=""

GCP_REGION=""
# Used for naming upload and output buckets
ORGANIZATION=""

hifisolves-ingest \
	--sample-info sample_info.csv \
	--backend gcp \
	--region "${GCP_REGION}" \
	--organization "${ORGANIZATION}"

Development

Tests

Note that you will need access to have the active cloud-specific credentials below set for the various cloud backends to run the tests.

See this secret for the values you'll need to set here.

# Required AWS credentials
export AWS_ACCESS_KEY_ID=""
export AWS_SECRET_ACCESS_KEY=""
export AWS_SESSION_TOKEN=""
# alternatively - just AWS_PROFILE
# export AWS_PROFILE=""

# Required Azure credentials
_Note that these credentials will eventually expire_
# R/A/W/C/D/L on destination container
export AZURE_STORAGE_SAS_TOKEN=""
# R/L on src container
export SOURCE_CONTAINER_SAS_TOKEN=""

# Path to service account JSON
export GOOGLE_APPLICATION_CREDENTIALS=""

python3 -m unittest discover -b -s tests

Setting the workflow version

The workflow version is comprised of two parts:

• WORKFLOW_VERSION: This is the version of the HumanWGS workflow in use; it should refer to a specific tagged version of this workflow
• WORKFLOW_SUB_VERSION: This is the revision of the HumanWGS workflow, and is used when we need to make changes to the workflow that are not present in PacBio's official workflow

Changing any part of either of these versions will force a new version of the workflow to be created in the user's namespace. Changing the WORKFLOW_VERSION or the major version of the WORKFLOW_SUB_VERSION will also result in the script resetting the run status for all samples back to unprocessed; changing just the minor or patch version of WORKFLOW_SUB_VERSION will allow any run with the same major WORKFLOW_SUB_VERSION to be picked up & used to determine run status. This allows small bugfixes to be made to the workflow in order to enable running samples without the need to fully reprocess all samples when a new version of the workflow is registered.

Setting workflow version automatically using a Git hook

There are two locations where the workflow version is set manually (in constants.py and in the hifisolves_wrapper workflow itself). This may lead to issues when there is a new release and fork of the HumanWGS pipeline and these values are not adjusted accordingly. Running the below will set the path to the Git hooks directory to our custom hooks directory and run a blank git checkout command, which invokes the post-checkout hook and sets the workflow version in both of the aforementioned locations.

Note: The command will need to be run from the root of the repository (hifi-solves-run-humanwgs)

git config core.hooksPath hooks/ && git checkout

Building Packages

A Makefile resides in the root directory of the package. Rules have been created to clean, build, release and push a package release. The following sections provide details on building and pushing the code to a repository

Building the package

make build

This creates a Python package to be uploaded to the Python Package Index (PyPI: pypi.org).

Building and pushing the Docker image

make docker-build

This rule builds a Docker image and pushes it to a container registry.