Sparse deconvolution of cell type medleys in spatial transcriptomics
File(s)
Published version
Author(s)
Sogunmez Erdogan, Nuray
Eroglu, Deniz
Type
Journal Article
Abstract
Mapping cell distributions across spatial locations with whole-genome coverage is essential for understanding cellular responses and signaling pathways. However, current deconvolution models, which aim to estimate the proportions of distinct cell types in each spatial transcriptomics spot by integrating reference single-cell data, often assume strong overlap between reference and spatial datasets, neglecting biology-grounded constraints like sparsity and cell-type variations and technical sparsity resulting in noise. As a result, these methods rely on over-permissive algorithms that ignore given constraints leading to inaccurate predictions, particularly in heterogeneous or unmatched datasets. We introduce Weight-Induced Sparse Regression (WISpR), a machine learning algorithm that integrates spot-specific hyperparameters and sparsity-driven modeling. Unlike conventional approaches that neglect biology-grounded constraints, WISpR accurately predicts cell-type distributions while preserving biological coherence, i.e., spatially and functionally consistent cell-type localization, even in unmatched datasets. Benchmarking against five alternative methods across ten datasets, WISpR consistently outperformed competitors and predicted cellular landscapes in both normal and cancerous tissues. By leveraging sparse cell-type arrangements, WISpR provides biologically informed, high-resolution cellular maps. Its ability to decode tissue organization in both healthy and diseased states highlights WISpR’s practical utility for spatial transcriptomics, particularly in challenging settings involving noise, sparsity, or reference mismatches.
Date Issued
2025-06-12
Date Acceptance
2025-05-27
Citation
PLoS Computational Biology, 2025, 21 (6)
ISSN
1553-734X
Publisher
Public Library of Science (PLoS)
Journal / Book Title
PLoS Computational Biology
Volume
21
Issue
6
Copyright Statement
© 2025 Erdogan, Eroglu. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
License URL
Publication Status
Published
Article Number
e1013169
Date Publish Online
2025-06-12