SDePER - A hybrid machine learning and regression method for cell-type deconvolution of spatial barcoding-based transcriptomic data

Aug 5, 2026

Upload by Noah

Spatial barcoding-based transcriptomic (ST) data need deconvolution to enable cellular-level downstream analysis. SDePER introduces a hybrid machine learning and regression framework that uses reference single-cell RNA sequencing (scRNA-seq) data. The method addresses platform effects between ST and scRNA-seq by enforcing a linear relationship, while also modeling sparsity and spatial correlations across capture spots. SDePER estimates cell-type proportions and improves tissue mapping by imputing cell-type compositions and gene expressions at unmeasured locations. Results on simulated and four real datasets show superior accuracy and robustness compared with existing approaches.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages30

Reading Time1 h 16 min

AccessFree to download

Abstract

Background

Spatial transcriptomic technologies

Need for cell-type deconvolution

SDePER Overview

The Texas Medical Center Library DigitalCommons@TMC | Faculty and Staff Publications | Baylor College of Medicine | | --- | --- | | 10-14-2024
SDePER: A Hybrid Machine Learning and Regression Method for Cell-Type Deconvolution of Spatial Barcoding-Based Transcriptomic Data
Yunqing Liu Ningshan Li Ji Qi Gang Xu Jiayi Zhao
Follow this and additional works at: [https://digitalcommons.library.tmc.edu/baylor_docs](https://digitalcommons.library.tmc.edu/baylor_docs)
SeePnaertoe fBoilaoditaiol ahleanuotoera, Cell Phenomena, and Immunity Commons, Critical Care Commons, Internal Medicine Commons, Medical Cell Biology Commons, Pulmonology Commons, and the Sleep Medicine Commons
| | Recommended Citation Liu, Yunqing; Li, Ningshan; Qi, Ji; Xu, Gang; Zhao, Jiayi; Wang, Nating; Huang, Xiayuan; Jiang, Wenhao; Wei, Huanhuan; Justet, Aurélien; Adams, Taylor S; Homer, Robert; Amei, Amei; Rosas, Ivan O; Kaminski, Naftali; Wang, Zuoheng; and Yan, Xiting, "SDePER: A Hybrid Machine Learning and Regression Method for CellType Deconvolution of Spatial Barcoding-Based Transcriptomic Data" (2024) . Faculty and Staff Publications. 1757. [https://digitalcommons.library.tmc.edu/baylor_docs/1757](https://digitalcommons.library.tmc.edu/baylor_docs/1757) This Article is brought to you for free and open access by the Baylor College of Medicine at DigitalCommons@TMC. It has been accepted for inclusion in Faculty and Staff Publications by an authorized administrator of DigitalCommons@TMC. For more information, please contact [digcommons@library.tmc.edu](digcommons@library.tmc.edu). 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, and Xiting Yan This article is available at DigitalCommons@TMC: [https://digitalcommons.library.tmc.edu/baylor_docs/1757](https://digitalcommons.library.tmc.edu/baylor_docs/1757) Liu et al. Genome Biology (2024) 25:271 [https://doi.org/10.1186/s13059-024-03416-2](https://doi.org/10.1186/s13059-024-03416-2) Genome Biology METHOD Open Access SDePER: a hybrid machine learning and regression method for cell-type deconvolution of spatial barcoding-based transcriptomic data Yunqing Liu1†, Ningshan Li1,2,3†, Ji Qi1, Gang Xu1,4, Jiayi Zhao1, Nating Wang1, Xiayuan Huang1, Wenhao Jiang1, Huanhuan Wei1,5, Aurélien Justet5,6, Taylor S. Adams5, Robert Homer7, Amei Amei4, Ivan O. Rosas8, Naftali Kaminski5, Zuoheng Wang1,9* and Xiting Yan1,5* | †Yunqing Liu and Ningshan Li contributed equally to this work. | | --- | | *Correspondence: [zuoheng.wang@yale.edu](zuoheng.wang@yale.edu); [xiting.yan@yale.edu](xiting.yan@yale.edu) | 1 Department of Biostatistics, Yale School of Public Health, New Haven, CT, USA 5 Section of Pulmonary, Critical Care and Sleep Medicine, Yale School of Medicine, New Haven, CT, USA Full list of author information is available at the end of the article Abstract Spatial barcoding-based transcriptomic (ST) data require deconvolution for cellularlevel 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 scRNAseq 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. Background Spatial transcriptomic technologies enabled measuring gene expression and physical locations of spots and/or cells simultaneously in intact tissues of various types inan unbiased and high-thro

FAQs

What is SDePER designed to do?

SDePER is a hybrid machine learning and regression method for deconvolving spatial barcoding-based transcriptomic data into cell-type compositions and cell-type-specific signals using reference scRNA-seq data.

How does SDePER handle differences between ST and scRNA-seq platforms?

SDePER tackles platform effects by ensuring a linear relationship between ST and scRNA-seq data while addressing sparsity and spatial correlations across capture spots.

What benefits does SDePER provide for tissue mapping?

By estimating cell-type proportions and imputing cell-type compositions and gene expressions at unmeasured locations, SDePER enables enhanced resolution tissue mapping.

Aug 5, 2026

Upload by Noah