Multiple sequence alignment
Alignments that start where homology search ends: built from ranked hits over object storage, deterministic column for column.
From hits to columns
One search, one alignment.
A multiple sequence alignment is homology search made columnar: the hits a query already found, stacked so that related residues share a column and every gap is explicit. Rafflesia treats it as a primitive of the same engine, not a separate pipeline to operate.
Search native
An alignment is derived from the same immutable releases homology search reads, so the rows in your profile are the hits you already ranked.
Deterministic
Same query, same release, same alignment. Columns, gaps, and coordinates reproduce byte for byte.
Provenance built in
Every row carries its sequence identity, coordinates, and the release it came from, so a profile is traceable, not a snapshot.
Rafflesia MSA
Profiles at corpus scale.
Bring a family, a profile workload, or a corpus you want to align against. We will help you compile it into a release and serve it.