Skip to content

pansift

A succinct index for population-scale pangenome graphs.

A compressed, queryable index over pangenome graphs containing thousands of haplotype paths. pansift keeps the full graph + path set in memory at roughly 18 bytes per node for vertebrate-scale graphs, and supports sub-millisecond path queries needed for graph-coordinate variant projection and graph-aware read alignment. Used internally as the index backend for graphmap2 and svhound.