2023Terashi DeepMainMast: Difference between revisions

From 3DEM-Methods
Jump to navigation Jump to search
(Created page with "== Citation == Cheng, Yuanhao / Huang, Xiaojun / Xu, Bin / Ding, Wei. AutoEMage: automatic data transfer, preprocessing, real-time display and monitoring in cryo-EM. 2023. J....")
 
No edit summary
 
Line 1: Line 1:
== Citation ==
== Citation ==


Cheng, Yuanhao / Huang, Xiaojun / Xu, Bin / Ding, Wei. AutoEMage: automatic data transfer, preprocessing, real-time display and monitoring in cryo-EM. 2023. J. Applied Crystallography, Vol. 56, No. 6, p. 1865-1873
Terashi, Genki / Wang, Xiao / Prasad, Devashish / Nakamura, Tsukasa / Kihara, Daisuke. DeepMainmast: integrated protocol of protein structure modeling for cryo-EM with deep learning and structure prediction. 2023. Nature Methods, p. 1-10


== Abstract ==
== Abstract ==


Cryo-electron microscopy (cryo-EM), especially single-particle analysis, has
Three-dimensional structure modeling from maps is an indispensable
become a powerful technique for visualizing the structure of biological
step for studying proteins and their complexes with cryogenic electron
macromolecules at high resolution. However, data acquisition in cryo-EM is
microscopy. Although the resolution of determined cryogenic electron
time consuming because it requires the collection of thousands of images to
microscopy maps has generally improved, there are still many cases where
achieve a high-quality reconstruction. Real-time preprocessing and display of
tracing protein main chains is difficult, even in maps determined at a
the images can greatly enhance the efficiency and quality of data collection. This
near-atomic resolution. Here we developed a protein structure modeling
study presents AutoEMage, a new open-source software package that automates
method, DeepMainmast, which employs deep learning to capture the
data transfer, preprocessing and real-time display in cryo-EM experiments.
local map features of amino acids and atoms to assist main-chain tracing.
AutoEMage also includes a real-time data monitoring system that alerts users to
Moreover, we integrated AlphaFold2 with the de novo density tracing
issues with their data, allowing them to take corrective actions accordingly. The
protocol to combine their complementary strengths and achieved even
software is equipped with an easy-to-use graphical user interface that provides
higher accuracy than each method alone. Additionally, the protocol is
seamless data screening and real-time feedback on data quality and microscope
able to accurately assign the chain identity to the structure models of
status.
homo-multimers, which is not a trivial task for existing methods.


== Keywords ==
== Keywords ==

Latest revision as of 08:12, 10 January 2024

Citation

Terashi, Genki / Wang, Xiao / Prasad, Devashish / Nakamura, Tsukasa / Kihara, Daisuke. DeepMainmast: integrated protocol of protein structure modeling for cryo-EM with deep learning and structure prediction. 2023. Nature Methods, p. 1-10

Abstract

Three-dimensional structure modeling from maps is an indispensable step for studying proteins and their complexes with cryogenic electron microscopy. Although the resolution of determined cryogenic electron microscopy maps has generally improved, there are still many cases where tracing protein main chains is difficult, even in maps determined at a near-atomic resolution. Here we developed a protein structure modeling method, DeepMainmast, which employs deep learning to capture the local map features of amino acids and atoms to assist main-chain tracing. Moreover, we integrated AlphaFold2 with the de novo density tracing protocol to combine their complementary strengths and achieved even higher accuracy than each method alone. Additionally, the protocol is able to accurately assign the chain identity to the structure models of homo-multimers, which is not a trivial task for existing methods.

Keywords

Links

https://www.nature.com/articles/s41592-023-02099-0

Related software

Related methods

Comments