Filters








20 Hits in 4.5 sec

Equivariant Graph Neural Networks for 3D Macromolecular Structure [article]

Bowen Jing, Stephan Eismann, Pratham N. Soni, Ron O. Dror
2021 arXiv   pre-print
Here, we extend recent work on geometric vector perceptrons and apply equivariant graph neural networks to a wide range of tasks from structural biology.  ...  Our method outperforms all reference architectures on three out of eight tasks in the ATOM3D benchmark, is tied for first on two others, and is competitive with equivariant networks using higher-order  ...  Conclusion We have demonstrated the systematic application of equivariant graph neural networks to macromolecular structures.  ... 
arXiv:2106.03843v2 fatcat:3uhjvnsrwjdyrghcb4c66c7vx4

Geometric Deep Learning on Molecular Representations [article]

Kenneth Atz, Francesca Grisoni, Gisbert Schneider
2021 arXiv   pre-print
Geometric deep learning (GDL), which is based on neural network architectures that incorporate and process symmetry information, has emerged as a recent paradigm in artificial intelligence.  ...  This review provides an overview of current challenges and opportunities, and presents a forecast of the future of GDL for molecular sciences.  ...  E(3)-equivariant message passing graph neural networks applied to three-dimensional (3D) molecular graphs: 3D graphs G 3 = (V, E, R) that are labeled with atom features (v i ∈ R dv ), their absolute coordinates  ... 
arXiv:2107.12375v4 fatcat:sgxlqdxiavbinly4s3zthysxbq

Protein sequence-to-structure learning: Is this the end(-to-end revolution)? [article]

Elodie Laine, Stephan Eismann, Arne Elofsson, Sergei Grudinin
2021 arXiv   pre-print
The potential of deep learning has been recognized in the protein structure prediction community for some time, and became indisputable after CASP13.  ...  architectures, i.e. differentiable models starting from a sequence and returning a 3D structure.  ...  The authors thank Kliment Olechnovič from Vilnius University for his help with illustrating Voronoi cells and proof-reading the manuscript, and Bowen Jing for his feedback on the manuscript.  ... 
arXiv:2105.07407v2 fatcat:6szubg7q2rajlj3l4vyzqri3nm

EGR: Equivariant Graph Refinement and Assessment of 3D Protein Complex Structures [article]

Alex Morehead, Xiao Chen, Tianqi Wu, Jian Liu, Jianlin Cheng
2022 arXiv   pre-print
In this work, we introduce the Equivariant Graph Refiner (EGR), a novel E(3)-equivariant graph neural network (GNN) for multi-task structure refinement and assessment of protein complexes.  ...  In doing so, we establish a baseline for future studies in macromolecular refinement and structure analysis.  ...  Our work follows that of [12] to incorporate E(3)-equivariance in our message passing neural network for 3D structure refinement and quality assessment.  ... 
arXiv:2205.10390v2 fatcat:peyeyv2t3rbifk5jx632a4g4pq

SegmA: Residue Segmentation of cryo-EM density maps [article]

Haim J. Wolfson, Mark Rozanov
2021 bioRxiv   pre-print
We present a deep neural network (nicknamed SegmA) for the residue type segmentation of a cryo-EM density map.  ...  The network labels voxels in a cryo-EM map by the residue type (amino acid type or nucleic acid) of the sampled macromolecular structure.  ...  Structure Generator (Li et al. (2020) ) uses CNN to estimate amino acids with their poses, which are then fed to a graph convolutional network and a recurrent network to obtain the 3D protein structure  ... 
doi:10.1101/2021.07.25.453685 fatcat:p3j7s2xvvfcwxmc4pcthaj2n5y

Protein structure prediction by AlphaFold2: are attention and symmetries all you need?

Nazim Bouatta, Peter Sorger, Mohammed AlQuraishi
2021 Acta Crystallographica Section D: Structural Biology  
The functions of most proteins result from their 3D structures, but determining their structures experimentally remains a challenge, despite steady advances in crystallography, NMR and single-particle  ...  Recent advances in deep learning, combined with the availability of genomic data for inferring co-evolutionary patterns, provide a new approach to protein structure prediction that is complementary to  ...  MA is a member of the SAB of FL2021-002, a Foresite Labs company, and consults for Interline Therapeutics.  ... 
doi:10.1107/s2059798321007531 pmid:34342271 pmcid:PMC8329862 fatcat:sam47cns4fhg3hgo273qoshlta

Learning from Protein Structure with Geometric Vector Perceptrons [article]

Bowen Jing, Stephan Eismann, Patricia Suriana, Raphael J.L. Townshend, Ron Dror
2021 arXiv   pre-print
Graph neural networks equipped with such layers are able to perform both geometric and relational reasoning on efficient and natural representations of macromolecular structure.  ...  Learning on 3D structures of large biomolecules is emerging as a distinct area in machine learning, but there has yet to emerge a unifying network architecture that simultaneously leverages the graph-structured  ...  Acknowledgements We thank Tri Dao and all members of the Dror group for feedback and discussions. Funding We acknowledge support from the U.S.  ... 
arXiv:2009.01411v3 fatcat:gn2xsyfvorfafa6lptfmv5fsji

AN ANALYSIS OF PROTEIN STRUCTURE PREDICTION WITH HELP OF ARTIFICIAL INTELLIGENCE

Aditi Bhadoriya, The Pacific International Public School
2022 International Journal of Engineering Applied Sciences and Technology  
Figuring out what shapes proteins fold into is known as ―protein folding problem‖ and stood as a grand challenge in biology for the past fifty years.  ...  advance, the latest version of the Artificial Intelligence system Alpha Fold has been recognized as a solution to this grand challenge by the organizers of the biennial Critical Assessment of Protein Structure  ...  For the final module, it employs a transformer in the system that updates the graph neural network with attention.  ... 
doi:10.33564/ijeast.2022.v07i04.027 fatcat:2xq35ntm2bhkrhsfaciwrtkcce

VoroCNN: Deep convolutional neural network built on 3D Voronoi tessellation of protein structures

Ilia Igashov, Liment Olechnovič, Maria Kadukova, Česlovas Venclovas, Sergei Grudinin
2021 Bioinformatics  
For the first time, we present a deep convolutional neural network (CNN) constructed on a Voronoi tessellation of 3D molecular structures.  ...  The prediction results are competitive to state of the art and superior to the previous 3D CNN architectures built for the same task.  ...  Acknowledgements The authors would like to thank Elodie Laine from Sorbonne Université for the discussions during the study and proof-reading the manuscript.  ... 
doi:10.1093/bioinformatics/btab118 pmid:33620450 fatcat:t4uxwxxo5vfg3lsjwttds7ur4e

DProQ: A Gated-Graph Transformer for Protein Complex Structure Assessment [article]

Xiao Chen, Alex Morehead, Jian Liu, Jianlin Cheng
2022 arXiv   pre-print
We challenge this significant task with DProQ, which introduces a gated neighborhood-modulating Graph Transformer (GGT) designed to predict the quality of 3D protein complex structures.  ...  Notably, we incorporate node and edge gates within a novel Graph Transformer framework to control information flow during graph message passing.  ...  Moreover, the Equivariant Graph Refiner (EGR) model formulated structural refinement and assessment of protein complexes using semi-supervised equivariant graph neural networks.  ... 
arXiv:2205.10627v1 fatcat:gwynnqykrvbdpa3zd6nizi3r3q

DISCA: high-throughput cryo-ET structural pattern mining by deep unsupervised clustering [article]

Xiangrui Zeng, Anson Kahng, Liang Xue, Julia Mahamid, Yi-Wei Chang, Min Xu
2021 bioRxiv   pre-print
We introduce a high-throughput template-and-label-free deep learning approach that automatically discovers subsets of homogeneous structures by learning and modeling 3D structural features and their distributions  ...  Diverse structures emerging from sorted subsets enable systematic unbiased recognition of macromolecular complexes in situ.  ...  Neural network architecture design A tomogram is a grayscale 3D volume of very large size (e.g., 4000×6000×1000 voxels).  ... 
doi:10.1101/2021.05.16.444381 fatcat:qul2vproqreexka46mttrri4tu

Simulate Time-integrated Coarse-grained Molecular Dynamics with Geometric Machine Learning [article]

Xiang Fu, Tian Xie, Nathan J. Rebello, Bradley D. Olsen, Tommi Jaakkola
2022 arXiv   pre-print
graph neural networks.  ...  Despite only trained with short MD trajectory data, our learned simulator can generalize to unseen novel systems and simulate for much longer than the training trajectories.  ...  2134795), and MIT-GIST collaboration for support.  ... 
arXiv:2204.10348v2 fatcat:z5kplawl45eb3pent5p5wrueti

Contrastive Representation Learning for 3D Protein Structures [article]

Pedro Hermosilla, Timo Ropinski
2022 arXiv   pre-print
To address this challenge, we introduce a new representation learning framework for 3D protein structures.  ...  Learning from 3D protein structures has gained wide interest in protein modeling and structural bioinformatics.  ...  Several authors represent the protein tertiary structure as a 3D density map, and process it with a 3D convolutional neural network (3DCNN).  ... 
arXiv:2205.15675v1 fatcat:uy5ldhbcmnbl3ke5h4c4eawkse

Multi-Scale Representation Learning on Proteins [article]

Vignesh Ram Somnath, Charlotte Bunne, Andreas Krause
2022 arXiv   pre-print
This paper introduces a multi-scale graph construction of a protein -- HoloProt -- connecting surface to structure and sequence.  ...  Our graph encoder then learns a multi-scale representation by allowing each level to integrate the encoding from level(s) below with the graph at that level.  ...  Moreover, we thank Mojmír Mutný and Clemens Isert for their valuable feedback.  ... 
arXiv:2204.02337v1 fatcat:v5mdbnyi6bcvzdujgh4uxrr4yi

Protein Representation Learning by Geometric Structure Pretraining [article]

Zuobai Zhang, Minghao Xu, Arian Jamasb, Vijil Chenthamarakshan, Aurelie Lozano, Payel Das, Jian Tang
2022 arXiv   pre-print
In this paper, we propose to pretrain protein representations according to their 3D structures. We first present a simple yet effective encoder to learn the geometric features of a protein.  ...  Despite the effectiveness of sequence-based approaches, the power of pretraining on known protein structures, which are available in smaller numbers only, has not been explored for protein property prediction  ...  ACKNOWLEDGMENTS The authors would like to thank Meng Qu, Zhaocheng Zhu, Shengchao Liu, Chence Shi, Minkai Xu and Huiyu Cai for their helpful discussions and comments.  ... 
arXiv:2203.06125v4 fatcat:qq644zivgzcu7o2wnc2qhxlco4
« Previous Showing results 1 — 15 out of 20 results