The huge protein database that spawned AlphaFold and biology's AI revolution
Briefly

The Protein Data Bank (PDB), which holds over 200,000 protein structures, is essential for AI tools like AlphaFold to accurately predict protein structures from sequences.
Helen Berman emphasizes the vital role of good data across scientific fields in facilitating AI breakthroughs, reflecting on the historical challenges of sharing protein structures.
Before the PDB, protein structures were shared via punch cards or magnetic tape, showcasing the need for a centralized database to streamline collaboration.
Berman's forward-thinking in the 1960s led to the creation of the PDB, emphasizing her belief in the impending significance of protein structures in science.
Read at Nature
[
|
]