- The human genome sequence is a composite from multiple individuals and does not capture actual human genetic variability.
- The FASTA file format, used for DNA storage, can only represent a single linear sequence, failing to encode population-level variations.
- A pan-genome captures genes present in some but not all individuals, as seen in bacterial studies like S. pneumoniae.
- Graph-based DNA storage allows representation of common and variable sequences, including optional elements and inversions.
- Efforts by Erik Garrison and the vg team focus on a full graph-aware software stack, though it requires more computational resources.
- Heng Li's rGFA format offers a simpler graph approach with stable coordinates, but is less generic and may lock in reference sequences.
- The shift to graph-based pan-genomes faces adoption hurdles due to convenience and existing linear DNA processing pipelines.