Systematic evaluation of robustness of deconvolution methods for spatial transcriptomics data in case of cell type mismatch
Briefings in Bioinformatics · 2026
DOI: 10.1093/bib/bbag594 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Utkarsh M Mahamune und 4 weiteren Autor:innen. Beteiligte Einrichtung: Epidemiology and Data Science, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105AZ Amsterdam, The Netherlands. Hauptthemen laut MEDLINE: Computational Biology, Gene Expression Profiling, Transcriptome. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: ARCAID, Marie Skłodowska-Curie.
Abstract (Originalsprache)
Sequencing-based spatial transcriptomics (ST) preserves spatial information but with limited cellular resolution, whereas single-cell RNA-sequencing (scRNA-seq) provides single-cell resolution but loses spatial context during tissue dissociation. Given these complementary strengths, computational tools combine scRNA-seq and ST data using deconvolution to identify cell types and estimate their proportions at each spatial location using scRNA-seq reference data. However, these methods are sensitive to missing cell types in the reference data, a problem known as cell type mismatch. Using two reference datasets, we performed extensive simulations to systematically evaluate the robustness to cell type mismatch of six deconvolution methods (CARD, cell2location, RCTD, Seurat, SPOTlight, and Stereoscope) tailored for ST data, and two designed for bulk RNA-seq data (MuSiC, SCDC). At baseline, i.e. with no cell types missing from the reference datasets, cell2location showed the strongest performance, while Seurat performed the worst. By simulating different cell type mismatch scenarios, we found that the performance of deconvolution methods decreases proportionally to the number of cell types missing from the reference. Moreover, compared to baseline, the relative decrease in performance is similar across methods and therefore no method is more robust to cell type mismatch than others. Additionally, methods that perform well at baseline tend to assign the proportions of a missing cell type to the transcriptionally most similar cell types present in the reference data. Our results highlight the adverse effects of cell type mismatch on the performance of deconvolution methods for ST data and stress the need for more robust approaches to this issue.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Epidemiology and Data Science, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105AZ Amsterdam, The Netherlands.
- Amsterdam Public Health, Methodology, Meibergdreef 9, 1105AZ Amsterdam, The Netherlands.
- Amsterdam institute for Immunology & Infectious Diseases, Inflammatory Diseases, Meibergdreef 9, 1105AZ Amsterdam, The Netherlands.
- Rheumatology and Clinical Immunology, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105AZ Amsterdam, The Netherlands.
- Amsterdam Rheumatology & Immunology Center (ARC), Meibergdreef 9, 1105AZ Amsterdam, The Netherlands.
ORCID-Profile: Utkarsh M Mahamune, Aldo Jongejan, Antoine H C van Kampen, Lisa G M van Baarsen, Perry D Moerland
Förderung
- ARCAID
- Marie Skłodowska-Curie
Zitieren
APA
Mahamune, U. M., Jongejan, A., van Kampen, A. H. C., van Baarsen, L. G. M., & Moerland, P. D. (2026). Systematic evaluation of robustness of deconvolution methods for spatial transcriptomics data in case of cell type mismatch. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag594
Vancouver
Mahamune UM, Jongejan A, van Kampen AHC, van Baarsen LGM, Moerland PD. Systematic evaluation of robustness of deconvolution methods for spatial transcriptomics data in case of cell type mismatch. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag594
BibTeX
@article{Mahamune2026,
author = {Mahamune, Utkarsh M and Jongejan, Aldo and van Kampen, Antoine H C and van Baarsen, Lisa G M and Moerland, Perry D},
title = {Systematic evaluation of robustness of deconvolution methods for spatial transcriptomics data in case of cell type mismatch},
journal = {Briefings in Bioinformatics},
year = {2026},
doi = {10.1093/bib/bbag594}
}
EUVIMED