CryoID2: deep learning-enhanced identification of protein and RNA from cryoEM maps
- 1University of California, Los Angeles
Abstract
Identifying the molecular components of cryoEM/cryoET maps of endogenous or poorly characterized complexes remains a major bottleneck. CryoID2 addresses this by integrating deep learning-based feature prediction with sequence prioritization and structural evaluation. Three 3D U-Nets extract structure classes, key-atom positions, and reduced residue/nucleotide types from the input map. These predictions support backbone tracing and the generation of short sequence queries that rank candidates from a user-supplied pool-using BLASTp for proteins and a maximin strategy for RNA. Corresponding PDB or AlphaFold models are then retrieved, rigidly fitted using predicted features and experimental density, and ranked by map agreement. Module-level tests show high-recall feature prediction and resolution-dependent atom clustering, while a 52-case benchmark recovers near-reference fitting quality. In practice, cryoID2 correctly identified PfA-M17 from a 166-protein mass-spectrometry pool and the unannotated Ca-raiA RNA from a genome-derived non-coding sequence set, demonstrating a unified sequence-plus-structure route to molecular identification in cryoEM maps.
Related
- Machine-learning-enhanced Optical Photothermal infrared (O-PTIR) spectroscopy for rapid simultaneous identification of multiple metabolites to monitor biotransformations: proof of concept on Brewer's spent grain (BSG)
- Integrated hierarchical fusion of GRU and meta-features for enhanced identification and interpretation of calcium ion inhibitors
- Variable-resolution maps (VRMs) in CCTBX and Phenix : accounting for local resolution in cryoEM
- Sample preparation for direct nanopore based protein identification via tryptic peptide recognition
- MgATP/MgADP-dependent conformational dynamics and intrinsically disordered regions of vascular KATP channels revealed by cryoEM
BibTeX
@misc{xu2026cryoid2,
title = {{CryoID2: deep learning-enhanced identification of protein and RNA from cryoEM maps}},
author = {Q. Xu and X. Yu and L. Wu and S. Feng and Z. H. Zhou},
howpublished = {bioRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.03.764069},
url = {https://www.biorxiv.org/content/10.64898/2026.11.03.764069v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - CryoID2: deep learning-enhanced identification of protein and RNA from cryoEM maps
AU - Q. Xu
AU - X. Yu
AU - L. Wu
AU - S. Feng
AU - Z. H. Zhou
PY - 2026
DA - 2026/11/03
PB - bioRxiv
DO - 10.64898/2026.11.03.764069
UR - https://www.biorxiv.org/content/10.64898/2026.11.03.764069v1
AB - Identifying the molecular components of cryoEM/cryoET maps of endogenous or poorly characterized complexes remains a major bottleneck. CryoID2 addresses this by integrating deep learning-based feature prediction with sequence prioritization and structural evaluation. Three 3D U-Nets extract structure classes, key-atom positions, and reduced residue/nucleotide types from the input map. These predictions support backbone tracing and the generation of short sequence queries that rank candidates from a user-supplied pool-using BLASTp for proteins and a maximin strategy for RNA. Corresponding PDB or AlphaFold models are then retrieved, rigidly fitted using predicted features and experimental density, and ranked by map agreement. Module-level tests show high-recall feature prediction and resolution-dependent atom clustering, while a 52-case benchmark recovers near-reference fitting quality. In practice, cryoID2 correctly identified PfA-M17 from a 166-protein mass-spectrometry pool and the unannotated Ca-raiA RNA from a genome-derived non-coding sequence set, demonstrating a unified sequence-plus-structure route to molecular identification in cryoEM maps.
ER -