For more: High-throughput Based Antibody Screening
In the search for new antibodies, computational methods provide a tightly controlled parallel screening of various biophysical features. Because the sequence and structure of the cognate antibody are unknown, antibody structural modeling is crucial for computational antibody design. Because antibody structure frameworks are so substantially conserved, antibody structure prediction is typically done using a template-based approach. Additionally, most techniques use a similar pipeline process. In order to model the framework region, one or more template structures taken from the Protein Data Bank (PDB) or a curated database are first chosen and utilized as a scaffold for the next phases. Next, it is predicted which way the VH and VL domains will align. The torsion angles of the side chains are then predicted using only the backbone information after modeling the CDR loops in the third step.