Structure search

The goal of the structure search is to annotate the given input protein sequences to experimentally resolved protein structures. For that we use the sequence similarity search method MMseqs2. For running MMseqs2, we use a custom search database, constructed from the ATOM records of all structures in the PDB. Here we preferred the biological assembly for each PDB entry if available.
The results from the structure search are pairings of input sequences and protein structures and are given to the next step the Alignment.