SDePER: A Hybrid Machine Learning and Regression Method for Cell-Type Deconvolution of Spatial Barcoding-Based Transcriptomic Data

Authors

Yunqing Liu
Ningshan Li
Ji Qi
Gang Xu
Jiayi Zhao
Nating Wang
Xiayuan Huang
Wenhao Jiang
Huanhuan Wei
Aurélien Justet
Taylor S Adams
Robert Homer
Amei Amei
Ivan O Rosas
Naftali Kaminski
Zuoheng Wang
Xiting Yan

Publication Date

10-14-2024

Journal

Genome Biology

DOI

10.1186/s13059-024-03416-2

PMID

39402626

PMCID

PMC11475911

PubMedCentral® Posted Date

10-14-2024

PubMedCentral® Full Text Version

Post-print

Published Open-Access

yes

Keywords

Machine Learning, Single-Cell Analysis, Transcriptome, Humans, Gene Expression Profiling, Sequence Analysis, RNA, Software, Animals, Regression Analysis, RNA-Seq

Abstract

Spatial barcoding-based transcriptomic (ST) data require deconvolution for cellular-level downstream analysis. Here we present SDePER, a hybrid machine learning and regression method to deconvolve ST data using reference single-cell RNA sequencing (scRNA-seq) data. SDePER tackles platform effects between ST and scRNA-seq data, ensuring a linear relationship between them while addressing sparsity and spatial correlations in cell types across capture spots. SDePER estimates cell-type proportions, enabling enhanced resolution tissue mapping by imputing cell-type compositions and gene expressions at unmeasured locations. Applications to simulated data and four real datasets showed SDePER's superior accuracy and robustness over existing methods.