CLI Reference#
Defaults keep probabilities and dosage in float64 using Blosc zstd level 7
with byte shuffle. For probabilities only, use --no-dosage. For dosage-only
float32 output, use --no-probabilities --float-dtype float32; this discards
uncertainty and haplotype probabilities and explicitly rounds dosage.
--compressor lz4 --compression-level 3 changes output compression, not the
input BGEN compression. See bgen2zarr for examples and tradeoffs.
bgen2zarr convert#
Convert BGEN genotype and phased haplotype probabilities to VCF Zarr.
Usage
bgen2zarr convert [OPTIONS] BGEN_PATH ZARR_PATH
Options
- --probabilities, --no-probabilities#
Retain genotype/haplotype probabilities (GP/HP).
- Default:
True
- --dosage, --no-dosage#
Retain expected alternate-allele dosage (DS).
- Default:
True
- --float-dtype <float_dtype>#
GP/HP/DS storage precision.
- Default:
'float64'- Options:
float64 | float32
- --compressor <compressor>#
Output Blosc codec.
- Default:
'zstd'- Options:
zstd | lz4
- --compression-level <compression_level>#
Output Blosc compression level.
- Default:
7
- --sample-ids <sample_ids>#
UTF-8 file with one sample ID per line, in BGEN sample order.
- --sample-file <sample_file>#
Oxford .sample file in BGEN sample order; overrides embedded IDs. Only IDs are imported, not phenotypes/covariates. Exclusive with –sample-ids.
- --sample-id-column <sample_id_column>#
Identifier column in –sample-file (default: first column). Use ID_2 explicitly when the first column contains family IDs.
- --dimension-source <dimension_source>#
Discover widths from actual ploidies (default) or declared bounds. Mutually exclusive with –max-ploidy.
- Options:
actual | declared
- --probability-model <probability_model>#
Discover phase models, or allocate explicit models and validate on decode.
- Default:
'auto'- Options:
auto | unphased | phased | mixed
- --max-ploidy <max_ploidy>#
Checked sample-ploidy bound; skips dimension discovery with an explicit –probability-model.
- -f, --force#
Force overwriting of existing directories
- -p, --worker-processes <worker_processes>#
Number of worker processes
- Default:
0
- -P, --progress, -Q, --no-progress#
Show progress bars (default: show)
- -v, --verbose#
Increase verbosity
- -l, --variants-chunk-size <variants_chunk_size>#
Chunk size in the variants dimension
- -w, --samples-chunk-size <samples_chunk_size>#
Chunk size in the samples dimension
- --zarr-format <zarr_format>#
Zarr format version of output (default: 2, or value of BIO2ZARR_DEFAULT_ZARR_FORMAT env var)
- Options:
2 | 3
Arguments
- BGEN_PATH#
Required argument
- ZARR_PATH#
Required argument