omics.co.in logo omics.co.in

13. Pan-Genome Graph Construction & Variation Graph Genotyping

⚡ Pipeline 13 of 16  ·  🧬 Assembly & Pangenomics

Transitioning from reference bias toward non-linear multi-assembly variation graphs using Minigraph-Cactus, indexing with vg, and mapping short reads via vg giraffe.

🔍 Ctrl + K
Pan-Genomics • Graph Genomes Authority: Human Pangenome Reference Consortium (HPRC)
Production Blueprint v4.5

13. Pan-Genome Graph Construction & Variation Graph Genotyping

Transitioning from reference bias toward non-linear multi-assembly variation graphs using Minigraph-Cactus, indexing with vg, and mapping short reads via vg giraffe.

Metric / RequirementSpecificationTarget Configuration
Core Engine / FrameworkMinigraph-Cactus + vg giraffe + vg callNextflow / Snakemake Compatible
Benchmark Runtime~6 – 10 hours32 CPU cores / 64 GB RAM
Standard Containerquay.io/comparative-genomics-toolkit/cactus:v2.7.2Docker • Singularity • Conda
Input → DeliverableMultiple FASTA Assemblies → GBZ Graph, Non-Biased Mapped GAM, Graph VCF
Workflow Chain: Multiple FASTAs → minigraph-cactus → vg autoindex → vg giraffe (bias-free mapping) → vg call
⚡

Step 1: Construct Pangenome Variation Graph with Minigraph-Cactus

Cactus Pangenome

Integrates dozens of phased telomere-to-telomere assemblies into a reference graph.

cactus-pangenome ./jobStore ./assemblies.seqfile \
                 --outDir ./pangenome_out \
                 --outName human_pangenome \
                 --reference GRCh38 \
                 --vcf --gbz --gfa
Key Flags & Notes: --gbz emits the Graph-Burrows-Wheeler compressed index; --vcf creates VCF representation of all graph bubbles.
⚡

Step 2: Generate vg Giraffe High-Performance Graph Indices

vg autoindex

Builds distance index, minimizer index, and snarl decompositions for rapid k-mer alignment.

vg autoindex --workflow giraffe -p pangenome_out/human_pangenome.gbz -g pangenome_index
Key Flags & Notes: Generates optimized fast indices used by vg giraffe for short-read graph routing.
⚡

Step 3: Align Short Reads to Pangenome Graph with vg Giraffe

vg giraffe

Aligns Illumina reads to the non-linear pangenome, completely eliminating human reference allele bias.

vg giraffe --gbz-name pangenome_out/human_pangenome.gbz \
           --dist-name pangenome_out/human_pangenome.dist \
           --min-name pangenome_out/human_pangenome.min \
           -f sample_R1.fq.gz -f sample_R2.fq.gz > sample_mapped.gam
Key Flags & Notes: Reads align across alternative structural alleles as easily as reference alleles.
⚡

Step 4: Graph-Based Genotype Calling with vg call

vg pack • vg call

Packs alignment coverage depths across graph nodes/edges and calls phased variants.

vg pack -x pangenome_out/human_pangenome.gbz -g sample_mapped.gam -Q 5 -o sample.pack
vg call pangenome_out/human_pangenome.gbz -k sample.pack > sample_graph_variants.vcf
Key Flags & Notes: Produces standard VCF callset referenced to the primary backbone coordinates.

🚀 Automated Nextflow (nf-core) Production Execution

Execute this complete workflow across any SLURM, AWS Batch, or Google Cloud HPC cluster using the verified community container pipeline:

# Orchestrated via Cactus container or Nextflow pangenome pipeline
← Return to Home Hub