EMReady2: improvement of cryo-EM and cryo-ET maps by local quality-aware deep learning with Mamba

Cryo-electron microscopy (cryo-EM) has emerged as a leading technology for determining the structures of biological macromolecules. However, map quality issues such as noise and loss of contrast hinder accurate map interpretation. Traditional and deep learning-based post-processing methods offer improvements but face limitations particularly in handling map heterogeneity. Here, we present EMReady2, an extension of our previous EMReady cryo-EM map improvement method. EMReady2 introduces a fast Mamba-based dual-branch UNet architecture to jointly capture local and global features. In addition, EMReady2 also uses a local resolution-guided learning strategy to address map local quality heterogeneity, and significantly extends the training set. These advances render EMReady2 applicable to a broader range of cryo-EM maps, including those containing nucleic acids, medium-resolution maps, and cryo-electron tomography (cryo-ET) maps. EMReady2 is evaluated on 136 diverse maps at 2.0-10.0 Å resolutions, and compared with existing map post-processing methods. Our results demonstrate that EMReady2 exhibits state-of-the-art performance in both map quality and map interpretability improvement while much reducing the computational cost.