gdata.GenomeDataBuilder#

class gdata.GenomeDataBuilder(self, /, location, genome_fasta, window_size, *, segments=None, step_size=None, chunk_size=None, resolution=1, chroms=None, overwrite=False)#

Represents a builder for genomic data, allowing for the creation and management of genomic datasets.

This struct provides methods to create a new genomic dataset, open an existing dataset, and manage genomic data chunks. It supports operations like adding files, retrieving chromosome information, and iterating over data chunks.

The builder creates a structured dataset in a specified location:

` root/ ├── metadata.json ├── chr1/ │   ├── 0/ │   │   ├── data |   |   ├── data.index │   │   ├── sequence.dat |   |   └── names.txt │   |── 1/ │   │   ├── data |   |   ├── data.index │   │   ├── sequence.dat |   |   └── names.txt `

Parameters:

  • location – The directory where the genomic data will be stored.

  • genome_fasta – The path to the FASTA file containing the genome sequences.

  • window_size – The size of the genomic windows to be processed.

  • segments – Optional list of genomic segments to include in the dataset. The genomic segments should be provided as strings in the format “chrom:start-end”. If None, the entire genome will be used.

  • step_size – The step size for sliding the window across the genome (default is None, which uses window_size).

  • chunk_size – The number of segments to store in each chunk. If None, it will be calculated based on the formula 2^(25 - log2(window_size)).

  • resolution – The resolution of the stored genomic data (default is 1).

  • chroms – A list of chromosomes to include in the dataset. If None, all chromosomes in the FASTA file will be used.

  • overwrite – If True, existing data at the specified location will be overwritten (default is False).

See also

GenomeDataLoader

Examples

>>> from gdata import as GenomeDataBuilder
>>> regions = ["chr11:35041782-35238390", "chr11:35200000-35300000"]
>>> tracks = {'DNase:CD14-positive monocyte': 'ENCSR464ETX.w5z', 'DNase:keratinocyte': 'ENCSR000EPQ.w5z'}
>>> builder = GenomeDataBuilder("genome", 'genome.fa.gz', segments=regions, window_size=196_608, resolution=128)
>>> builder.add_files(tracks)

Methods

add_file(key, w5z)

Adds a single file to the dataset.

add_files(files)

Adds w5z files to the dataset.

chroms()

Returns a vector of chromosome names present in the genomic data.

get_chunk_size()

Returns the size of each chunk in the dataset.

open()

Open an existing GenomeDataBuilder instance from a specified location.

segments()

Returns the segments of the genome as a vector of strings.

tracks()

Returns the keys (track names) in the dataset.
On this page