Title: Multi-head attention-based U-Nets for predicting protein domain boundaries using 1D sequence features and 2D distance maps

Journal Article · · BMC Bioinformatics
; ; ; ;

Abstract The information about the domain architecture of proteins is useful for studying protein structure and function. However, accurate prediction of protein domain boundaries (i.e., sequence regions separating two domains) from sequence remains a significant challenge. In this work, we develop a deep learning method based on multi-head U-Nets (called DistDom) to predict protein domain boundaries utilizing 1D sequence features and predicted 2D inter-residue distance map as input. The 1D features contain the evolutionary and physicochemical information of protein sequences, whereas the 2D distance map includes the structural information of proteins that was rarely used in domain boundary prediction before. The 1D and 2D features are processed by the 1D and 2D U-Nets respectively to generate hidden features. The hidden features are then used by the multi-head attention to predict the probability of each residue of a protein being in a domain boundary, leveraging both local and global information in the features. The residue-level domain boundary predictions can be used to classify proteins as single-domain or multi-domain proteins. It classifies the CASP14 single-domain and multi-domain targets at the accuracy of 75.9%, 13.28% more accurate than the state-of-the-art method. Tested on the CASP14 multi-domain protein targets with expert annotated domain boundaries, the average per-target F1 measure score of the domain boundary prediction by DistDom is 0.263, 29.56% higher than the state-of-the-art method.

Research Organization:
Donald Danforth Plant Science Center, St. Louis, MO (United States); University of Missouri, Columbia, MO (United States)
Sponsoring Organization:
National Institutes of Health (NIH); National Science Foundation (NSF); USDOE; USDOE Advanced Research Projects Agency - Energy (ARPA-E); USDOE Office of Science (SC); USDOE Office of Science (SC), Biological and Environmental Research (BER)
Grant/Contract Number:
AC05-00OR22725; AR0001213; SC0020400; SC0021303
OSTI ID:
1876657
Journal Information:
BMC Bioinformatics, Journal Name: BMC Bioinformatics Journal Issue: 1 Vol. 23; ISSN 1471-2105
Publisher:
Springer Science + Business MediaCopyright Statement
Country of Publication:
United Kingdom
Language:
English

References (39)

CHOP proteins into structural domain-like fragments journal April 2004
PPRODO: Prediction of protein domain boundaries using neural networks journal March 2005
Analysis of distance‐based protein structure prediction by deep learning in CASP13 journal August 2019
DNSS2 : Improved ab initio protein secondary structure prediction using advanced deep learning architectures journal September 2020
Prediction of protein assemblies, the next frontier: The CASP14‐CAPRI experiment journal September 2021
Target highlights in CASP14 : Analysis of models by structure providers journal October 2021
SnapDRAGON: a method to delineate protein structural domains from sequence data journal February 2002
U-Net: Convolutional Networks for Biomedical Image Segmentation
  • Ronneberger, Olaf; Fischer, Philipp; Brox, Thomas
  • Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III https://doi.org/10.1007/978-3-319-24574-4_28
book November 2015
DomSVR: domain boundary prediction with support vector regression from sequence information alone journal February 2010
PDP-CON: prediction of domain/linker residues in protein sequences using a consensus approach journal March 2016
DOMpro: Protein Domain Prediction Using Profiles, Secondary Structure, Relative Solvent Accessibility, and Recursive Neural Networks journal May 2006
TopDomain: Exhaustive Protein Domain Boundary Metaprediction Combining Multisource Information and Deep Learning journal June 2021
Expression screening, protein purification and NMR analysis of human protein domains for structural genomics journal March 2004
Improved protein structure prediction using potentials from deep learning journal January 2020
Artificial intelligence in the prediction of protein–ligand interactions: recent advances and future directions journal November 2021
FUpred: detecting protein domains through deep-learning-based contact map prediction journal March 2020
A deep dilated convolutional residual network for predicting interchain contacts of protein homodimers journal February 2022
SSEP-Domain: protein domain prediction by alignment of secondary structure elements and profiles journal November 2005
DROP: an SVM domain linker predictor trained with optimal features selected by random forest journal December 2010
ThreaDom: extracting protein domain boundary information from multiple threading alignments journal June 2013
ConDo: protein domain boundary prediction using coevolutionary information journal November 2018
DNN-Dom: predicting protein domain boundary from sequence alone by deep neural network journal June 2019
Gapped BLAST and PSI-BLAST: a new generation of protein database search programs journal September 1997
Protein structure prediction servers at University College London journal July 2005
KemaDom: a web server for domain prediction using kernel machine with local context journal July 2006
ImageNet: A large-scale hierarchical image database
  • Deng, Jia; Dong, Wei; Socher, Richard
  • 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2009 IEEE Conference on Computer Vision and Pattern Recognition https://doi.org/10.1109/CVPR.2009.5206848
conference June 2009
High-Performance Deep Learning Toolbox for Genome-Scale Prediction of Protein Structure and Function conference November 2021
Computer-aided NMR assay for detecting natively folded structural domains journal March 2006
DeepDom: Predicting protein domain boundary from sequence alone using stacked bidirectional LSTM conference November 2018
The Natural History of Protein Domains journal June 2002
Long Short-Term Memory journal November 1997
DoBo: Protein domain boundary prediction by integrating evolutionary signals and machine learning journal February 2011
Improving the performance of DomainDiscovery of protein domain boundary assignment using inter-domain linker index journal December 2006
Improved general regression network for protein domain boundary prediction journal February 2008
Propedia: a database for protein–peptide identification based on a hybrid clustering algorithm journal January 2021
A multi-source domain annotation pipeline for quantitative metagenomic and metatranscriptomic functional profiling journal August 2018
Improvement in Protein Domain Identification Is Reached by Breaking Consensus, with the Agreement of Many Profiles and Domain Co-occurrence journal July 2016
DomHR: Accurately Identifying Domain Boundaries in Proteins Using a Hinge Region Strategy journal April 2013
Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned conference January 2019

Similar Records

Related Subjects

59 BASIC BIOLOGICAL SCIENCES
attention
deep learning
protein distance map
protein distance map
protein domain boundary
protein sequence