Implementation of a 3D structure similarity search for proteins and multimeric assemblies based on a novel embedding model
[
ref].
This application is a prototype; the official implementation will be integrated into the
RCSB.org search.
The dataset includes protein chains and assemblies from RCSB PDB and AlphaFold DB.
Data content was frozen in December 2024; newer PDB or AlphaFold DB structures are not available.
The current implementation runs on a CPU-only instance. Querying by a coordinates file may take some time, depending on the structure size.