Full-length<i>de novo</i>protein structure determination from cryo-EM maps using deep learning

He, Jiahua; Huang, Sheng‐You

doi:10.1101/2020.08.28.271981

Cited by 4 publications

(3 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To facilitate the second structure recognition, machine learning approaches are employed to automatically detect secondary structures as implemented in programs such as SSELearner 6 , γ-TEMPy 7 , EMNUSS 8 , Emap2sec 9 . Recently, deep learning (DL) methods based on 3D convolutional neural networks, such as UNet 10 , were also employed to predict protein backbone structure 11,12 or model protein complex structures [13][14][15] .…”

Section: Introductionmentioning

confidence: 99%

Integrating AlphaFold and deep learning for atomistic interpretation of cryo-EM maps

Dai

Yoo

et al. 2023

Preprint

View full text Add to dashboard Cite

Interpretation of cryo-electron microscopy (cryo-EM) maps requires building and fitting 3-D atomic models of biological molecules. AlphaFold-predicted models generate initial 3-D coordinates; however, model inaccuracy and conformational heterogeneity often necessitate labor-intensive manual model building and fitting into cryo-EM maps. In this work, we designed a protein model-building workflow, which combines a deep-learning cryo-EM map enhancement tool, ResEM(Resolution EnhanceMent) and AlphaFold. A benchmark test using 37 cryo-EM maps shows that ResEM achieves state-of-the-art performance in optimizing real space model-map correlations between ResEM-enhanced maps and ground-truth models. Furthermore, in a subset of 17 datasets where the initial AlphaFold predictions are less accurate, the workflow significantly improves their model accuracy. Our work demonstrates that the integration of modern deep learning image enhancement and AlphaFold may lead to automated model building and fitting for the atomistic interpretation of cryo-EM maps.

show abstract

Section: Introductionmentioning

confidence: 99%

Integrating AlphaFold and deep learning for atomistic interpretation of cryo-EM maps

Dai

Yoo

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…At these resolutions, model building is time consuming, error prone, and often ambiguous. To assist this process, methods have been developed to automatically build de novo polypeptide chains into EM data [4][5][6][7], and with the advent of AlphaFold 2, high-quality starting models can oftentimes be obtained from sequence information alone [8,9]. While these methods help build protein models into cryoEM density, tools for automatic fitting of small molecule ligands into cryoEM data are limited.…”

Section: Introductionmentioning

confidence: 99%

Automatic and accurate ligand structure determination guided by cryo-electron microscopy maps

Muenks

Zepeda

Zhou

et al. 2022

Preprint

View full text Add to dashboard Cite

Advances in cryo-electron microscopy (cryoEM) and deep-learning guided protein structure prediction have expedited structural studies of protein complexes. However, methods for accurately determining ligand conformations are lacking. In this manuscript, we develop a tool for automatically determining ligand structures guided by medium-resolution cryoEM density. We show this method is robust at predicting ligands in maps as low as 6Å resolution, and is able to correct receptor sidechain errors. Combining this with a measure of placement confidence, and running on all protein/ligand structures in EMDB, we show that 58% of ligands replicate the deposited model, 16% confidently find alternate conformations, 22% have ambiguous density where multiple conformations might be present, and 4% are incorrectly placed. For five cases where our approach finds an alternate conformation with high confidence, high-resolution crystal structures validate our placement. This tool and the resulting analysis should prove critical in using cryoEM to investigate protein-ligand complexes.

show abstract

“…A related problem, de novo chain modeling, has been addressed by leveraging segmentation and symmetry [ 10 ]. More recently, the use of machine learning techniques has been applied in this space [ 11 , 12 , 13 , 14 ] and deep learning, in particular, has shown its potential to identify secondary structure elements in EM maps [ 15 , 16 , 17 , 18 ]. It is possible to use hybrid approaches to extract additional information from electron microscopy maps at resolutions up to 10 Å [ 19 ].…”

Section: Introductionmentioning

confidence: 99%

EvoSeg: Automated Electron Microscopy Segmentation through Random Forests and Evolutionary Optimization

Zumbado-Corrales

Esquivel‐Rodríguez

2021

Biomimetics

View full text Add to dashboard Cite

Electron Microscopy Maps are key in the study of bio-molecular structures, ranging from borderline atomic level to the sub-cellular range. These maps describe the envelopes that cover possibly a very large number of proteins that form molecular machines within the cell. Within those envelopes, we are interested to find what regions correspond to specific proteins so that we can understand how they function, and design drugs that can enhance or suppress a process that they are involved in, along with other experimental purposes. A classic approach by which we can begin the exploration of map regions is to apply a segmentation algorithm. This yields a mask where each voxel in 3D space is assigned an identifier that maps it to a segment; an ideal segmentation would map each segment to one protein unit, which is rarely the case. In this work, we present a method that uses bio-inspired optimization, through an Evolutionary-Optimized Segmentation algorithm, to iteratively improve upon baseline segments obtained from a classical approach, called watershed segmentation. The cost function used by the evolutionary optimization is based on an ideal segmentation classifier trained as part of this development, which uses basic structural information available to scientists, such as the number of expected units, volume and topology. We show that a basic initial segmentation with the additional information allows our evolutionary method to find better segmentation results, compared to the baseline generated by the watershed.

show abstract

Full-lengthde novoprotein structure determination from cryo-EM maps using deep learning

Cited by 4 publications

References 61 publications

Integrating AlphaFold and deep learning for atomistic interpretation of cryo-EM maps

Integrating AlphaFold and deep learning for atomistic interpretation of cryo-EM maps

Automatic and accurate ligand structure determination guided by cryo-electron microscopy maps

EvoSeg: Automated Electron Microscopy Segmentation through Random Forests and Evolutionary Optimization

Contact Info

Product

Resources

About