TY - JOUR
T1 - Complex hierarchical structures in single-cell genomics data unveiled by deep hyperbolic manifold learning
AU - Tian, Tian
AU - Zhong, Cheng
AU - Lin, Xiang
AU - Wei, Zhi
AU - Hakonarson, Hakon
N1 - Funding Information:
We thank Dr. Yao Ma from the New Jersey Institute of Technology for valuable suggestions. We thank Dr. Ruihua Cheng from the Tianjin University of Finance and Economics for proofreading, which improved the clarity of the manuscript. This work used the Extreme Science and Engineering Discovery Environment (XSEDE) through the allocation CIE170034, supported by National Science Foundation grant number ACI-1548562. This work was supported by grant R15HG012087 (Z.W.) from the National Institutes of Health (NIH) and was partially supported by the National Center for Advancing Translational Sciences (NCATS), a component of NIH under award number UL1TR003017 (Z.W.).
Publisher Copyright:
© 2023 Tian et al.
PY - 2023
Y1 - 2023
N2 - With the advances in single-cell sequencing techniques, numerous analytical methods have been developed for delineating cell development. However, most are based on Euclidean space, which would distort the complex hierarchical structure of cell differentiation. Recently, methods acting on hyperbolic space have been proposed to visualize hierarchical structures in single-cell RNA-seq (scRNA-seq) data and have been proven to be superior to methods acting on Euclidean space. However, these methods have fundamental limitations and are not optimized for the highly sparse single-cell count data. To address these limitations, we propose scDHMap, a model-based deep learning approach to visualize the complex hierarchical structures of scRNA-seq data in low-dimensional hyperbolic space. The evaluations on extensive simulation and real experiments show that scDHMap outperforms existing dimensionality-reduction methods in various common analytical tasks as needed for scRNA-seq data, including revealing trajectory branches, batch correction, and denoising the count matrix with high dropout rates. In addition, we extend scDHMap to visualize single-cell ATAC-seq data.
AB - With the advances in single-cell sequencing techniques, numerous analytical methods have been developed for delineating cell development. However, most are based on Euclidean space, which would distort the complex hierarchical structure of cell differentiation. Recently, methods acting on hyperbolic space have been proposed to visualize hierarchical structures in single-cell RNA-seq (scRNA-seq) data and have been proven to be superior to methods acting on Euclidean space. However, these methods have fundamental limitations and are not optimized for the highly sparse single-cell count data. To address these limitations, we propose scDHMap, a model-based deep learning approach to visualize the complex hierarchical structures of scRNA-seq data in low-dimensional hyperbolic space. The evaluations on extensive simulation and real experiments show that scDHMap outperforms existing dimensionality-reduction methods in various common analytical tasks as needed for scRNA-seq data, including revealing trajectory branches, batch correction, and denoising the count matrix with high dropout rates. In addition, we extend scDHMap to visualize single-cell ATAC-seq data.
UR - http://www.scopus.com/inward/record.url?scp=85149999013&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85149999013&partnerID=8YFLogxK
U2 - 10.1101/gr.277068.122
DO - 10.1101/gr.277068.122
M3 - Article
C2 - 36849204
AN - SCOPUS:85149999013
SN - 1088-9051
VL - 33
SP - 232
EP - 246
JO - Genome Research
JF - Genome Research
IS - 2
ER -