Genomes and annotations¶

Omnomnomics stores installed genome resources in a dedicated reference directory, separate from the software installation and individual experiment directories. Each assembly uses a normalized layout:

REFERENCE_ROOT/
├── assemblies/
│   └── GRCh38/
│       ├── fasta/genome.fa
│       ├── annotation/genes.gtf
│       ├── hisat2/
│       ├── star/
│       └── aux/
└── motif_databases/

Common operations:

omnomnomics genomes list --species human
omnomnomics genomes installed
omnomnomics genomes install --species mouse
omnomnomics genomes blacklist --assembly GRCh38
omnomnomics genomes motifs

genomes install normalizes FASTA and GTF locations and builds requested HISAT2/STAR indexes directly. STAR indexing can require substantially more memory than HISAT2 indexing.

When available, ENCODE blacklist files are cached in the assembly aux/ directory. The default JASPAR MEME database is cached once under motif_databases/; use genomes motifs --force to refresh it. For strict reproducibility, set an explicit MEME-format motif database in the workflow configuration.