AI tool outperforms existing x-ray structure methods

Birefringence

Source: © Dirk Wiersma/Science Photo Library

Deep learning approach determines phase of diffracted beams with lower-quality data

Scientists in Denmark have used artificial intelligence (AI) to resolve a key challenge in crystal-structure determination for the first time, and have made the system freely available for other researchers to use. The University of Copenhagen’s Anders Madsen and his colleagues have developed a deep-learning system called PhAI that works out hard-to-determine information about the phase of x-rays that crystals have diffracted.