scvi.data.add_dna_sequence#

scvi.data.add_dna_sequence(adata, seq_len=1344, genome_name='hg38', genome_dir=None, genome_provider=None, install_genome=True, chr_var_key='chr', start_var_key='start', end_var_key='end', sequence_varm_key='dna_sequence', code_varm_key='dna_code')[source]#

Add the DNA sequence to AnnData object.

Uses genomepy under the hood to download the genome.

Parameters:
adata AnnData

AnnData object with chromatin accessibility data

seq_len int (default: 1344)

Length of the DNA sequence to extract around the peak center. Defaults to value used in scBasset.

genome_name str (default: 'hg38')

Name of genome to use, installed with genomepy

genome_dir Path | None (default: None)

Directory to install genome to, if not already installed

genome_provider str | None (default: None)

Provider of genome, passed to genomepy

install_genome bool (default: True)

Install the genome with genomepy. If False, genome_provider is not used, and a genome is loaded with genomepy.Genome(genome_name, genomes_dir=genome_dir)

chr_var_key str (default: 'chr')

Key in .var for chromosome

start_var_key str (default: 'start')

Key in .var for start position

end_var_key str (default: 'end')

Key in .var for end position

sequence_varm_key str (default: 'dna_sequence')

Key in .varm for the added DNA sequence

code_varm_key str (default: 'dna_code')

Key in .varm for the added DNA sequence, encoded as integers

Return type:

None

Returns:

None

Adds fields to .varm:

sequence_varm_key: DNA sequence code_varm_key: DNA sequence, encoded as integers