TY - GEN
T1 - Adaptive total-variation joint learning model for analyzing single cell RNA seq data
AU - Zhang, Dai Jun
AU - Zhao, Jing Xiu
AU - Liu, Jin Xing
AU - Gao, Ying Lian
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - An important purpose of single-cell RNA sequencing (scRNA-seq) data research is to explain the complex and diverse heterogeneity information between cells, which can further deepen human understanding of the mechanisms of life and the organization of organisms. However, the high dimensionality and noise are two major factors that hinder the development of scRNA-seq data mining. Therefore, in this paper, an adaptive total-variant joint learning model (JL-ATV) is proposed to overcome these two drawbacks of scRNA-seq data mining. On the one hand, in this model, dimensionality reduction learning and segmentation reconstruction subspace methods is combined to obtain effective features descriptions of the scRNAseq data and improve the interpretability and accuracy of cell identification. On the other hand, a gradient-based learning approach, namely adaptive total variation (ATV), is applied to scRNA-seq data to preserve the internal structure and overcome the interference of noise. Finally, experiments on multiple datasets show that the JL-ATV model can obtain a set of effective features and further improve the accuracy of identifying cell types.
AB - An important purpose of single-cell RNA sequencing (scRNA-seq) data research is to explain the complex and diverse heterogeneity information between cells, which can further deepen human understanding of the mechanisms of life and the organization of organisms. However, the high dimensionality and noise are two major factors that hinder the development of scRNA-seq data mining. Therefore, in this paper, an adaptive total-variant joint learning model (JL-ATV) is proposed to overcome these two drawbacks of scRNA-seq data mining. On the one hand, in this model, dimensionality reduction learning and segmentation reconstruction subspace methods is combined to obtain effective features descriptions of the scRNAseq data and improve the interpretability and accuracy of cell identification. On the other hand, a gradient-based learning approach, namely adaptive total variation (ATV), is applied to scRNA-seq data to preserve the internal structure and overcome the interference of noise. Finally, experiments on multiple datasets show that the JL-ATV model can obtain a set of effective features and further improve the accuracy of identifying cell types.
KW - Cell clustering
KW - Dimensionality reduction
KW - Joint learning
KW - Single-cell RNA sequencing data
KW - Sparse representation
UR - https://www.scopus.com/pages/publications/85125185331
U2 - 10.1109/BIBM52615.2021.9669379
DO - 10.1109/BIBM52615.2021.9669379
M3 - 会议稿件
AN - SCOPUS:85125185331
T3 - Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021
SP - 775
EP - 778
BT - Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021
A2 - Huang, Yufei
A2 - Kurgan, Lukasz
A2 - Luo, Feng
A2 - Hu, Xiaohua Tony
A2 - Chen, Yidong
A2 - Dougherty, Edward
A2 - Kloczkowski, Andrzej
A2 - Li, Yaohang
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021
Y2 - 9 December 2021 through 12 December 2021
ER -