Reference Data

ST-CNVBench uses two kinds of reference data.

Bundled Small References

These files are already included in git under refs/hg38_genome_info/:

• hg38_genes_simple.txt
• hg38_genes_annot.txt
• cytoBand.txt.gz
• hg38.list

Population Phasing Bundle

Large population phasing resources are required only for allele-aware wrappers:

• CalicoST
• Numbat
• Xclone

Download the bundle from:

• Population phasing reference bundle (Google Drive)

After download, extract the bundle under:

refs/
└── population_phasing/

Required Files

The extracted population_phasing/ directory must contain:

• 1000G_hg38/
• genome1K.phase3.SNP_AF5e2.chr1toX.hg38.ensemble_style.sorted.vcf.gz
• genome1K.phase3.SNP_AF5e2.chr1toX.hg38.ensemble_style.sorted.vcf.gz.tbi

Expected layout:

refs/
└── population_phasing/
    ├── 1000G_hg38/
    ├── genome1K.phase3.SNP_AF5e2.chr1toX.hg38.ensemble_style.sorted.vcf.gz
    └── genome1K.phase3.SNP_AF5e2.chr1toX.hg38.ensemble_style.sorted.vcf.gz.tbi

Do not rename the files or the directory.

Practical Rule

If you are only running the shipped demo with CopyKAT, you do not need the population phasing bundle.