TY - JOUR
T1 - An analytical framework for interpretable and generalizable single-cell data analysis
AU - Zhou, Jian
AU - Troyanskaya, Olga G.
N1 - Funding Information:
The authors acknowledge all members of the Troyanskaya laboratory and Zhou laboratory for helpful discussions. This work was performed using the high-performance computing resources (supported by the Scientific Computing Core) at the Flatiron Institute and the BioHPC at UT Southwestern Medical Center. J.Z. is supported by the Cancer Prevention and Research Institute of Texas grant (RR190071) and the UT Southwestern Endowed Scholars program. O.G.T. is supported by National Institutes of Health grant nos. R01HG005998, U54HL117798 and R01GM071966, US Department of Health and Human Services grant no. HHSN272201000054C and Simons Foundation grant no. 395506. O.G.T. is a senior fellow of the Genetic Networks program of the Canadian Institute for Advanced Research.
Publisher Copyright:
© 2021, The Author(s), under exclusive licence to Springer Nature America, Inc.
PY - 2021/11
Y1 - 2021/11
N2 - The scaling of single-cell data exploratory analysis with the rapidly growing diversity and quantity of single-cell omics datasets demands more interpretable and robust data representation that is generalizable across datasets. Here, we have developed a ‘linearly interpretable’ framework that combines the interpretability and transferability of linear methods with the representational power of non-linear methods. Within this framework we introduce a data representation and visualization method, GraphDR, and a structure discovery method, StructDR, that unifies cluster, trajectory and surface estimation and enables their confidence set inference.
AB - The scaling of single-cell data exploratory analysis with the rapidly growing diversity and quantity of single-cell omics datasets demands more interpretable and robust data representation that is generalizable across datasets. Here, we have developed a ‘linearly interpretable’ framework that combines the interpretability and transferability of linear methods with the representational power of non-linear methods. Within this framework we introduce a data representation and visualization method, GraphDR, and a structure discovery method, StructDR, that unifies cluster, trajectory and surface estimation and enables their confidence set inference.
UR - http://www.scopus.com/inward/record.url?scp=85118366425&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85118366425&partnerID=8YFLogxK
U2 - 10.1038/s41592-021-01286-1
DO - 10.1038/s41592-021-01286-1
M3 - Article
C2 - 34725480
AN - SCOPUS:85118366425
SN - 1548-7091
VL - 18
SP - 1317
EP - 1321
JO - Nature Methods
JF - Nature Methods
IS - 11
ER -