PhAI: A deep-learning approach to solve the crystallographic phase problem

• 成果类型: Article
• 署名作者: Larsen, Anders S.; Rekis, Toms; Madsen, Anders O.
• 署名单位: University of Copenhagen
• 刊物名称: SCIENCE
• ISSN/ISSBN: 0036-13442
• DOI: 10.1126/science.adn2777
• 发表日期: 2024-08-02
• 页码: 522-528
• 关键词: charge-flipping algorithm; initio structure solution; crystal-structures; database; density; form

摘要

X-ray crystallography provides a distinctive view on the three-dimensional structure of crystals. To reconstruct the electron density map, the complex structure factors F = |F|exp(i phi) of a sufficiently large number of diffracted reflections must be known. In a conventional experiment, only the amplitudes F are obtained, and the phases phi are lost. This is the crystallographic phase problem. In this work, we show that a neural network, trained on millions of artificial structure data, can solve the phase problem at a resolution of only 2 angstroms, using only 10 to 20% of the data needed for direct methods. The network works in common space groups and for modest unit-cell dimensions and suggests that neural networks could be used to solve the phase problem in the general case for weakly scattering crystals.