DiffGAN: a conditional generative adversarial network for phasing single molecule diffraction data to atomic resolution

Abstract

IntroductionProteins that adopt multiple conformations pose significant challenges in structural biology research and pharmaceutical development, as structure determination via single particle cryo-electron microscopy (cryo-EM) is often impeded by data heterogeneity. In this context, the enhanced signal-to-noise ratio of single molecule cryo-electron diffraction (simED) offers a promising alternative. However, a significant challenge in diffraction methods is the loss of phase information, which is crucial for accurate structure determination. MethodsHere, we present DiffGAN, a conditional generative adversarial network (cGAN) that estimates the missing phases at high resolution from a combination of high-resolution single particle diffraction data and low-resolution image data. ResultsFor simulated datasets, DiffGAN allows effectively determine protein structures at atomic resolution from diffraction patterns and noisy low-resolution images. DiscussionOur findings suggest that combining single particle cryo-electron diffraction with advanced generative modeling, as in DiffGAN, could revolutionize the way protein structures are determined, offering a more accurate and efficient alternative to existing methods.