UMINT-FS: UMINT-guided Feature Selection for multi-omics datasets

Chayan Maitra; Dibyendu B. Seal; Vivek Das; Yevgeniy Vorobeychik; Rajat K. De

doi:10.1109/BIBM58861.2023.10385731

UMINT-FS: UMINT-guided Feature Selection for multi-omics datasets

Chayan Maitra
, Dibyendu B. Seal
, Vivek Das
, Yevgeniy Vorobeychik
, Rajat K. De

Department of Computer Science & Engineering

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

Abstract

Feature selection is a crucial step in single-cell biological data analysis. It involves identifying and selecting a subset of features (genes, proteins, peaks among others) that are most informative and relevant for downstream analysis. A prior investigation has introduced an unsupervised neural network model, known as UMINT, tailored for the integration of single-cell multi-omics data. This novel deep learning model excels at single-cell multi-omics integration and feature extraction, yet lacks the ability to perform feature selection. The present study extends UMINT and introduces UMINT-FS that enables selection of top features from multi-omics datasets by analysing the weights learned by the UMINT network during integration of the omics modalities. UMINT-FS can operate in both supervised and unsupervised learning environments. A supervised learning environment empowers it to find cell-type-specific markers. The performance of UMINT-FS has been evaluated on two different types of single-cell multi-omics datasets and results demonstrated better performance than current state-of-the-art methods.

Original language	English
Title of host publication	Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023
Editors	Xingpeng Jiang, Haiying Wang, Reda Alhajj, Xiaohua Hu, Felix Engel, Mufti Mahmud, Nadia Pisanti, Xuefeng Cui, Hong Song
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	594-601
Number of pages	8
ISBN (Electronic)	9798350337488
DOIs	https://doi.org/10.1109/BIBM58861.2023.10385731
State	Published - 2023
Event	2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023 - Istanbul, Turkey Duration: Dec 5 2023 → Dec 8 2023

Publication series

Name	Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

Conference

Conference	2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023
Country/Territory	Turkey
City	Istanbul
Period	12/5/23 → 12/8/23

Keywords

ATAC-seq
CITE-seq
Deep learning
Feature selection
Single-cell multi-omics integration

Access to Document

10.1109/BIBM58861.2023.10385731

Cite this

Maitra, C., Seal, D. B., Das, V., Vorobeychik, Y., & De, R. K. (2023). UMINT-FS: UMINT-guided Feature Selection for multi-omics datasets. In X. Jiang, H. Wang, R. Alhajj, X. Hu, F. Engel, M. Mahmud, N. Pisanti, X. Cui, & H. Song (Eds.), Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023 (pp. 594-601). (Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BIBM58861.2023.10385731

Maitra, Chayan ; Seal, Dibyendu B. ; Das, Vivek et al. / UMINT-FS : UMINT-guided Feature Selection for multi-omics datasets. Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023. editor / Xingpeng Jiang ; Haiying Wang ; Reda Alhajj ; Xiaohua Hu ; Felix Engel ; Mufti Mahmud ; Nadia Pisanti ; Xuefeng Cui ; Hong Song. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 594-601 (Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023).

@inproceedings{8cbfb464db674233987506fa4c122ce2,

title = "UMINT-FS: UMINT-guided Feature Selection for multi-omics datasets",

abstract = "Feature selection is a crucial step in single-cell biological data analysis. It involves identifying and selecting a subset of features (genes, proteins, peaks among others) that are most informative and relevant for downstream analysis. A prior investigation has introduced an unsupervised neural network model, known as UMINT, tailored for the integration of single-cell multi-omics data. This novel deep learning model excels at single-cell multi-omics integration and feature extraction, yet lacks the ability to perform feature selection. The present study extends UMINT and introduces UMINT-FS that enables selection of top features from multi-omics datasets by analysing the weights learned by the UMINT network during integration of the omics modalities. UMINT-FS can operate in both supervised and unsupervised learning environments. A supervised learning environment empowers it to find cell-type-specific markers. The performance of UMINT-FS has been evaluated on two different types of single-cell multi-omics datasets and results demonstrated better performance than current state-of-the-art methods.",

keywords = "ATAC-seq, CITE-seq, Deep learning, Feature selection, Single-cell multi-omics integration",

author = "Chayan Maitra and Seal, \{Dibyendu B.\} and Vivek Das and Yevgeniy Vorobeychik and De, \{Rajat K.\}",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023 ; Conference date: 05-12-2023 Through 08-12-2023",

year = "2023",

doi = "10.1109/BIBM58861.2023.10385731",

language = "English",

series = "Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023",

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

pages = "594--601",

editor = "Xingpeng Jiang and Haiying Wang and Reda Alhajj and Xiaohua Hu and Felix Engel and Mufti Mahmud and Nadia Pisanti and Xuefeng Cui and Hong Song",

booktitle = "Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023",

}

Maitra, C, Seal, DB, Das, V, Vorobeychik, Y & De, RK 2023, UMINT-FS: UMINT-guided Feature Selection for multi-omics datasets. in X Jiang, H Wang, R Alhajj, X Hu, F Engel, M Mahmud, N Pisanti, X Cui & H Song (eds), Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023. Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023, Institute of Electrical and Electronics Engineers Inc., pp. 594-601, 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023, Istanbul, Turkey, 12/5/23. https://doi.org/10.1109/BIBM58861.2023.10385731

UMINT-FS: UMINT-guided Feature Selection for multi-omics datasets. / Maitra, Chayan; Seal, Dibyendu B.; Das, Vivek et al.
Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023. ed. / Xingpeng Jiang; Haiying Wang; Reda Alhajj; Xiaohua Hu; Felix Engel; Mufti Mahmud; Nadia Pisanti; Xuefeng Cui; Hong Song. Institute of Electrical and Electronics Engineers Inc., 2023. p. 594-601 (Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023).

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

TY - GEN

T1 - UMINT-FS

T2 - 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

AU - Maitra, Chayan

AU - Seal, Dibyendu B.

AU - Das, Vivek

AU - Vorobeychik, Yevgeniy

AU - De, Rajat K.

PY - 2023

Y1 - 2023

N2 - Feature selection is a crucial step in single-cell biological data analysis. It involves identifying and selecting a subset of features (genes, proteins, peaks among others) that are most informative and relevant for downstream analysis. A prior investigation has introduced an unsupervised neural network model, known as UMINT, tailored for the integration of single-cell multi-omics data. This novel deep learning model excels at single-cell multi-omics integration and feature extraction, yet lacks the ability to perform feature selection. The present study extends UMINT and introduces UMINT-FS that enables selection of top features from multi-omics datasets by analysing the weights learned by the UMINT network during integration of the omics modalities. UMINT-FS can operate in both supervised and unsupervised learning environments. A supervised learning environment empowers it to find cell-type-specific markers. The performance of UMINT-FS has been evaluated on two different types of single-cell multi-omics datasets and results demonstrated better performance than current state-of-the-art methods.

AB - Feature selection is a crucial step in single-cell biological data analysis. It involves identifying and selecting a subset of features (genes, proteins, peaks among others) that are most informative and relevant for downstream analysis. A prior investigation has introduced an unsupervised neural network model, known as UMINT, tailored for the integration of single-cell multi-omics data. This novel deep learning model excels at single-cell multi-omics integration and feature extraction, yet lacks the ability to perform feature selection. The present study extends UMINT and introduces UMINT-FS that enables selection of top features from multi-omics datasets by analysing the weights learned by the UMINT network during integration of the omics modalities. UMINT-FS can operate in both supervised and unsupervised learning environments. A supervised learning environment empowers it to find cell-type-specific markers. The performance of UMINT-FS has been evaluated on two different types of single-cell multi-omics datasets and results demonstrated better performance than current state-of-the-art methods.

KW - ATAC-seq

KW - CITE-seq

KW - Deep learning

KW - Feature selection

KW - Single-cell multi-omics integration

UR - https://www.scopus.com/pages/publications/85184861008

U2 - 10.1109/BIBM58861.2023.10385731

DO - 10.1109/BIBM58861.2023.10385731

M3 - Conference contribution

AN - SCOPUS:85184861008

T3 - Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

SP - 594

EP - 601

BT - Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

A2 - Jiang, Xingpeng

A2 - Wang, Haiying

A2 - Alhajj, Reda

A2 - Hu, Xiaohua

A2 - Engel, Felix

A2 - Mahmud, Mufti

A2 - Pisanti, Nadia

A2 - Cui, Xuefeng

A2 - Song, Hong

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 5 December 2023 through 8 December 2023

ER -

Maitra C, Seal DB, Das V, Vorobeychik Y, De RK. UMINT-FS: UMINT-guided Feature Selection for multi-omics datasets. In Jiang X, Wang H, Alhajj R, Hu X, Engel F, Mahmud M, Pisanti N, Cui X, Song H, editors, Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 594-601. (Proceedings - 2023 2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023). doi: 10.1109/BIBM58861.2023.10385731

UMINT-FS: UMINT-guided Feature Selection for multi-omics datasets

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this