DeepMind AI Solves Structures of 200 Million Proteins
Google's DeepMind expands its AlphaFold database to include over 200 million protein structures, covering nearly every known sequenced organism. This massive breakthrough replaces slow, expensive traditional methods with rapid AI predictions.
Google's DeepMind AI system, known as AlphaFold, successfully decodes the structures of over 200 million proteins, covering virtually every protein known to science. By predicting these complex 3D shapes from simple amino-acid sequences, the program provides an incredible asset to life sciences and medicine. Previously, scientists only manage to unravel the structures of a tiny fraction of these essential cellular building blocks.
Predicting how a linear chain of amino acids folds into a final 3D shape is notoriously difficult due to the staggering number of possible configurations. Traditional methods like X-ray crystallography and electron microscopy are highly accurate but extremely slow and expensive. Artificial intelligence is perfectly suited to sift through these unfathomable amounts of possibilities, turning a near-impossible puzzle into a solvable task.
After initially stunning the scientific community in 2020, DeepMind now expands its public database in collaboration with the European Molecular Biology Laboratory. The updated catalog covers proteins from almost every organism on Earth that has a sequenced genome. This enormous expansion makes structural biology data instantly accessible to researchers worldwide, accelerating future medical and biological discoveries.