AlphaFold

AlphaFold's triangle update and the Edge Transformer share the form of "updating an edge through a third vertex" but lack gauge constraints.

AlphaFold-type tools can be used to predict three-dimensional protein structures, generative models can design entirely new molecules or antibody sequences, and machine learning can be used for virtual screening.

The life sciences AI track includes protein structure prediction represented by DeepMind's AlphaFold series, computational drug discovery represented by Schrödinger and Insilico Medicine, and general research assistants, among other categories.

ProGen goes in the opposite direction from AlphaFold: given a target function (such as antimicrobial activity), it generates entirely new protein sequences that fold into the corresponding structures, forming a complementary "prediction-generation" loop.

In October 2024, the Nobel Prize in Chemistry was awarded to Demis Hassabis and John Jumper (Google DeepMind) and David Baker (University of Washington) for AlphaFold's breakthrough in protein structure prediction.

AlphaFold-predicted structures come with confidence metrics (such as pLDDT), and low-confidence regions often correspond to flexible loops; using them directly for docking easily produces false positives, and experimental validation remains a necessary step to confirm the conformation of key binding sites.

AlphaFold's success depended on the hundreds of dedicated TPUs owned by DeepMind, protein sequence databases accumulated over many years, and a top-tier interdisciplinary team, a configuration that is almost impossible to replicate in a single university laboratory.

AlphaFold was developed by DeepMind and achieved accuracy comparable to experimental methods in the 2020 CASP14 competition, and is regarded as one of the most significant breakthroughs in structural biology in fifty years.

There are already precedents for foundation models in science, such as DeepMind's AlphaFold making a breakthrough in protein structure prediction, Huawei's Pangu weather model surpassing traditional numerical methods in weather forecasting, and NASA and Microsoft collaborating to build Earth Copilot for Earth observation.

AlphaFold lowered the barrier to early target assessment and structure-based drug design, but it provides structural hypotheses rather than biological validation; predicted structures still require experimental confirmation and cannot replace functional experiments and preclinical toxicology assessments.