Automated model building and protein identification in cryo-EM maps

2024/02/26 by Kiarash Jamali, Lukas Käll, Rui Zhang +3 · 652 citations
Biochemistry, Genetics and Molecular Biology · Materials Science · #Biology #Biophysics #Computational biology #Computer science #Cryo-electron microscopy #Data science #Ecology #Enzyme Structure and Function #Genomics and Phylogenetic Studies #Glycosylation and Glycoproteins Research #Identification (biology)

paper · pdf · doi:10.1038/s41586-024-07215-4

published in Nature 628(8007), 450-457 (Nature Portfolio)

openalex publication_date 2024/02/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06

Abstract

Abstract Interpreting electron cryo-microscopy (cryo-EM) maps with atomic models requires high levels of expertise and labour-intensive manual intervention in three-dimensional computer graphics programs 1,2 . Here we present ModelAngelo, a machine-learning approach for automated atomic model building in cryo-EM maps. By combining information from the cryo-EM map with information from protein sequence and structure in a single graph neural network, ModelAngelo builds atomic models for proteins that are of similar quality to those generated by human experts. For nucleotides, ModelAngelo builds backbones with similar accuracy to those built by humans. By using its predicted amino acid probabilities for each residue in hidden Markov model sequence searches, ModelAngelo outperforms human experts in the identification of proteins with unknown sequences. ModelAngelo will therefore remove bottlenecks and increase objectivity in cryo-EM structure determination.

Cited by

• Structural Characterization of Native RNA Polymerase II Transcription Complexes and Nucleosomes in Drosophila melanogaster
• High-resolution in situ structures of hantavirus glycoprotein tetramers
• Naturally ornate RNA-only complexes revealed by cryo-EM
• Structural basis for the conformational protection of nitrogenase from O2
• SPACA9 and MNMIP1 bridge the seam of spermatid manchette microtubules
• Structural basis for curvature generation and functional specialization in spirochete flagella
• Structure-based design of soluble prefusion-stabilized herpes simplex virus type 2 glycoprotein B antigens
• From Possibility to Precision in Macromolecular Ensemble Prediction
• Structure of the ATP-driven methyl-coenzyme M reductase activation complex
• Distinct structural features of Pseudomonas aeruginosa ATP synthase revealed by cryo-electron microscopy
• AI-guided prefusion stabilization of the human coronavirus OC43 spike protein enables universal embecovirus antigen design
• Numerous rRNA molecules form the apicomplexan mitoribosome via repurposed protein and RNA elements
• Visualizing the modification landscape of the human 60S ribosomal subunit at close to atomic resolution
• Structure of the plant plastid-encoded RNA polymerase
• Apicomplexan mitoribosome from highly fragmented rRNAs to a functional machine
• Identification and structural characterization of anthrax toxin receptor 2 as the Clostridium perfringens NetF receptor
• Structure of the Methanosarcina mazei Mtr complex bound to the oxygen-stress responsive small protein MtrI
• Anthrax toxin receptor 2 is the Receptor for Clostridium perfringens NetF: Structural Insights into Toxin Binding and Pore Formation
• Architecture of the 8 MDa Hdr–Vhu–Fwd super-assembly in class I methanogens
• HURP facilitates spindle assembly by stabilizing microtubules and working synergistically with TPX2
• The herpes simplex origin-binding protein: mechanisms for sequence-specific DNA binding and dimerization revealed by Cryo-EM
• Connecting the dots: deep learning-based automated model building methods in cryo-EM
• A Surfactant Cocktail Overcomes Air-Water Interface Artifacts in Single-Particle CryoEM
• Structural diversity of axonemes across mammalian motile cilia
• Structural adaptations for enhanced translation kinetics in evolved ribosomes
• Capturing ribosomal structures in cellular extracts with cryoPRISM: A purification-free cryoEM approach reveals novel structural states
• Q -score as a reliability measure for protein, nucleic acid and small-molecule atomic coordinate models derived from 3DEM maps
• Mapping the ultrastructural topology of the corynebacterial cell surface
• Mapping the ultrastructural topology of the corynebacterial cell surface
• Identification of a seasonal influenza vaccine–induced broadly protective neuraminidase antibody
• Structure of an Archaeal Ribosome with a Divergent Active Site
• Protein-templated synthesis of dinucleotide repeat DNA by an antiphage reverse transcriptase
• Scalable protein design using optimization in a relaxed sequence space
• Atkinsonella hypoxylon virus capsid structure highlights the diversity of capsid proteins among the Partitiviridae
• Structures of the human adult muscle-type nicotinic receptor in resting and desensitized states
• AlphaFold as a prior: experimental structure determination conditioned on a pretrained neural network
• A labeled dataset for AI-based cryo-EM map enhancement
• In-cell architecture of the mitochondrial respiratory chain
• PERC: a suite of software tools for the curation of cryoEM data with application to simulation, modelling and machine learning
• DomainFit: Identification of protein domains in cryo-EM maps at intermediate resolution using AlphaFold2-predicted models
• Towards better structural models from cryo‐electron microscopy data with physics‐based methods
• Isolation of genetically diverse influenza antibodies highlights the role of IG germline gene variation and informs the design of population-comprehensive vaccine strategies
• Structural switch in acetylcholine receptors in developing muscle
• Protein Graph Neural Networks for Heterogeneous Cryo-EM Reconstruction
• Structural specializations of the sperm tail. [europepmc]
• EMDB-the Electron Microscopy Data Bank. [europepmc]
• Targeted protein degradation via intramolecular bivalent glues. [europepmc]
• Structural basis of Integrator-dependent RNA polymerase II termination. [europepmc]
• Structural basis for human Ca v 3.2 inhibition by selective antagonists. [europepmc]
• DomainFit: Identification of protein domains in cryo-EM maps at intermediate resolution using AlphaFold2-predicted models. [europepmc]
• Accurate Prediction of Protein Structural Flexibility by Deep Learning Integrating Intricate Atomic Structures and Cryo-EM Density Information. [europepmc]
• Structure of the red-shifted Fittonia albivenis photosystem I. [europepmc]
• Structural switch in acetylcholine receptors in developing muscle. [europepmc]
• An image processing pipeline for electron cryo-tomography in RELION-5. [europepmc]
• Structural basis of the bacterial flagellar motor rotational switching. [europepmc]
• Alzheimer's disease seeded tau forms paired helical filaments yet lacks seeding potential. [europepmc]
• Heteromeric amyloid filaments of ANXA11 and TDP-43 in FTLD-TDP type C. [europepmc]
• Structure of the Nipah virus polymerase phosphoprotein complex. [europepmc]
• Protein complex structure modeling by cross-modal alignment between cryo-EM maps and protein sequences. [europepmc]
• DiffModeler: large macromolecular structure modeling for cryo-EM maps using a diffusion model. [europepmc]
• Scalable protein design using optimization in a relaxed sequence space. [europepmc]
• A disease resistance protein triggers oligomerization of its NLR helper into a hexameric resistosome to mediate innate immunity. [europepmc]
• Structure of the Nipah virus polymerase complex. [europepmc]
• Structural diversity of axonemes across mammalian motile cilia. [europepmc]
• AI-based methods for biomolecular structure modeling for Cryo-EM. [europepmc]
• G-quadruplex-stalled eukaryotic replisome structure reveals helical inchworm DNA translocation. [europepmc]
• Autoimmune mechanisms elucidated through muscle acetylcholine receptor structures. [europepmc]
• Structural basis for the midnolin-proteasome pathway and its role in suppressing myeloma. [europepmc]
• Structure-guided engineering of type I-F CASTs for targeted gene insertion in human cells. [europepmc]

Related

• A Graph Neural Network Approach to Automated Model Building in Cryo-EM Maps
• Sequence-guided protein structure determination using graph convolutional and recurrent networks
• Beyond Current Boundaries: Integrating Deep Learning and AlphaFold for Enhanced Protein Structure Prediction from Low-Resolution Cryo-EM Maps
• Connecting the dots: deep learning-based automated model building methods in cryo-EM
• Neural Representations of Cryo-EM Maps and a Graph-Based Interpretation