Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification

Jing Gao; Fa Zhang; Kai Hu; Xuefeng Cui

doi:10.1109/BIBM55620.2022.9995701

Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification

Jing Gao^*
, Fa Zhang^*
, Kai Hu^*
, Xuefeng Cui^*

^*Corresponding author for this work

School of Medical Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

6 Citations (Scopus)

Abstract

Recent advances in spatial transcriptomics have enabled the comprehensive measurement of transcriptional profiles while retaining the spatial contextual information. Identifying spatial domains is a critical step in the analysis of spatially resolved transcriptomics. Existing unsupervised methods perform poorly on this task owing to the large amount of noise and dropout events in the transcriptomic profiles. To address this problem, we first extend an unsupervised algorithm to a supervised learning method that can identify useful features and reduce noise hindrance. Second, inspired by the classical convolution in convolutional neural networks (CNNs), we designed a regular hexagonal convolution to compensate for the missing gene expression patterns from adjacent nodes. Compared with the graph convolution in graph neural networks (GNNs), our hexagonal convolution can preserve the relative spatial location information of different nodes in graph-structured data. Third, based on the hexagonal convolution, a novel hexagonal Convolutional Neural Network (hexCNN) is proposed for spatial transcriptomics classification. Finally, we compared the proposed hexCNN with existing methods on the DLPFC dataset. The results show that hexCNN achieves a classification accuracy of 87.2% and an average Rand index (ARI) of 78.2% (1.9% and 3.3% higher than those of GNNs).

Original language	English
Title of host publication	Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022
Editors	Donald Adjeroh, Qi Long, Xinghua Shi, Fei Guo, Xiaohua Hu, Srinivas Aluru, Giri Narasimhan, Jianxin Wang, Mingon Kang, Ananda M. Mondal, Jin Liu
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	200-205
Number of pages	6
ISBN (Electronic)	9781665468190
DOIs	https://doi.org/10.1109/BIBM55620.2022.9995701
Publication status	Published - 2022
Event	2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022 - Las Vegas, United States Duration: 6 Dec 2022 → 8 Dec 2022

Publication series

Name	Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

Conference

Conference	2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022
Country/Territory	United States
City	Las Vegas
Period	6/12/22 → 8/12/22

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Convolutional neural network
Graph neural network
Spatial domain identification
Spatial transcriptomics
Supervised learning

Access to Document

10.1109/BIBM55620.2022.9995701

Cite this

Gao, J., Zhang, F., Hu, K., & Cui, X. (2022). Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification. In D. Adjeroh, Q. Long, X. Shi, F. Guo, X. Hu, S. Aluru, G. Narasimhan, J. Wang, M. Kang, A. M. Mondal, & J. Liu (Eds.), Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022 (pp. 200-205). (Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BIBM55620.2022.9995701

Gao, Jing ; Zhang, Fa ; Hu, Kai et al. / Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification. Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022. editor / Donald Adjeroh ; Qi Long ; Xinghua Shi ; Fei Guo ; Xiaohua Hu ; Srinivas Aluru ; Giri Narasimhan ; Jianxin Wang ; Mingon Kang ; Ananda M. Mondal ; Jin Liu. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 200-205 (Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022).

@inproceedings{033aa90ecc894bf984aad106aef3a66b,

title = "Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification",

abstract = "Recent advances in spatial transcriptomics have enabled the comprehensive measurement of transcriptional profiles while retaining the spatial contextual information. Identifying spatial domains is a critical step in the analysis of spatially resolved transcriptomics. Existing unsupervised methods perform poorly on this task owing to the large amount of noise and dropout events in the transcriptomic profiles. To address this problem, we first extend an unsupervised algorithm to a supervised learning method that can identify useful features and reduce noise hindrance. Second, inspired by the classical convolution in convolutional neural networks (CNNs), we designed a regular hexagonal convolution to compensate for the missing gene expression patterns from adjacent nodes. Compared with the graph convolution in graph neural networks (GNNs), our hexagonal convolution can preserve the relative spatial location information of different nodes in graph-structured data. Third, based on the hexagonal convolution, a novel hexagonal Convolutional Neural Network (hexCNN) is proposed for spatial transcriptomics classification. Finally, we compared the proposed hexCNN with existing methods on the DLPFC dataset. The results show that hexCNN achieves a classification accuracy of 87.2\% and an average Rand index (ARI) of 78.2\% (1.9\% and 3.3\% higher than those of GNNs).",

keywords = "Convolutional neural network, Graph neural network, Spatial domain identification, Spatial transcriptomics, Supervised learning",

author = "Jing Gao and Fa Zhang and Kai Hu and Xuefeng Cui",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022 ; Conference date: 06-12-2022 Through 08-12-2022",

year = "2022",

doi = "10.1109/BIBM55620.2022.9995701",

language = "English",

series = "Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "200--205",

editor = "Donald Adjeroh and Qi Long and Xinghua Shi and Fei Guo and Xiaohua Hu and Srinivas Aluru and Giri Narasimhan and Jianxin Wang and Mingon Kang and Mondal, \{Ananda M.\} and Jin Liu",

booktitle = "Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022",

address = "United States",

}

Gao, J, Zhang, F, Hu, K & Cui, X 2022, Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification. in D Adjeroh, Q Long, X Shi, F Guo, X Hu, S Aluru, G Narasimhan, J Wang, M Kang, AM Mondal & J Liu (eds), Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022. Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022, Institute of Electrical and Electronics Engineers Inc., pp. 200-205, 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022, Las Vegas, United States, 6/12/22. https://doi.org/10.1109/BIBM55620.2022.9995701

Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification. / Gao, Jing; Zhang, Fa; Hu, Kai et al.
Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022. ed. / Donald Adjeroh; Qi Long; Xinghua Shi; Fei Guo; Xiaohua Hu; Srinivas Aluru; Giri Narasimhan; Jianxin Wang; Mingon Kang; Ananda M. Mondal; Jin Liu. Institute of Electrical and Electronics Engineers Inc., 2022. p. 200-205 (Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification

AU - Gao, Jing

AU - Zhang, Fa

AU - Hu, Kai

AU - Cui, Xuefeng

PY - 2022

Y1 - 2022

N2 - Recent advances in spatial transcriptomics have enabled the comprehensive measurement of transcriptional profiles while retaining the spatial contextual information. Identifying spatial domains is a critical step in the analysis of spatially resolved transcriptomics. Existing unsupervised methods perform poorly on this task owing to the large amount of noise and dropout events in the transcriptomic profiles. To address this problem, we first extend an unsupervised algorithm to a supervised learning method that can identify useful features and reduce noise hindrance. Second, inspired by the classical convolution in convolutional neural networks (CNNs), we designed a regular hexagonal convolution to compensate for the missing gene expression patterns from adjacent nodes. Compared with the graph convolution in graph neural networks (GNNs), our hexagonal convolution can preserve the relative spatial location information of different nodes in graph-structured data. Third, based on the hexagonal convolution, a novel hexagonal Convolutional Neural Network (hexCNN) is proposed for spatial transcriptomics classification. Finally, we compared the proposed hexCNN with existing methods on the DLPFC dataset. The results show that hexCNN achieves a classification accuracy of 87.2% and an average Rand index (ARI) of 78.2% (1.9% and 3.3% higher than those of GNNs).

AB - Recent advances in spatial transcriptomics have enabled the comprehensive measurement of transcriptional profiles while retaining the spatial contextual information. Identifying spatial domains is a critical step in the analysis of spatially resolved transcriptomics. Existing unsupervised methods perform poorly on this task owing to the large amount of noise and dropout events in the transcriptomic profiles. To address this problem, we first extend an unsupervised algorithm to a supervised learning method that can identify useful features and reduce noise hindrance. Second, inspired by the classical convolution in convolutional neural networks (CNNs), we designed a regular hexagonal convolution to compensate for the missing gene expression patterns from adjacent nodes. Compared with the graph convolution in graph neural networks (GNNs), our hexagonal convolution can preserve the relative spatial location information of different nodes in graph-structured data. Third, based on the hexagonal convolution, a novel hexagonal Convolutional Neural Network (hexCNN) is proposed for spatial transcriptomics classification. Finally, we compared the proposed hexCNN with existing methods on the DLPFC dataset. The results show that hexCNN achieves a classification accuracy of 87.2% and an average Rand index (ARI) of 78.2% (1.9% and 3.3% higher than those of GNNs).

KW - Convolutional neural network

KW - Graph neural network

KW - Spatial domain identification

KW - Spatial transcriptomics

KW - Supervised learning

UR - https://www.scopus.com/pages/publications/85146654481

U2 - 10.1109/BIBM55620.2022.9995701

DO - 10.1109/BIBM55620.2022.9995701

M3 - Conference contribution

AN - SCOPUS:85146654481

T3 - Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

SP - 200

EP - 205

BT - Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

A2 - Adjeroh, Donald

A2 - Long, Qi

A2 - Shi, Xinghua

A2 - Guo, Fei

A2 - Hu, Xiaohua

A2 - Aluru, Srinivas

A2 - Narasimhan, Giri

A2 - Wang, Jianxin

A2 - Kang, Mingon

A2 - Mondal, Ananda M.

A2 - Liu, Jin

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

Y2 - 6 December 2022 through 8 December 2022

ER -

Gao J, Zhang F, Hu K, Cui X. Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification. In Adjeroh D, Long Q, Shi X, Guo F, Hu X, Aluru S, Narasimhan G, Wang J, Kang M, Mondal AM, Liu J, editors, Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 200-205. (Proceedings - 2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022). doi: 10.1109/BIBM55620.2022.9995701

Hexagonal Convolutional Neural Network for Spatial Transcriptomics Classification

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this