CausalCellInfer: Prioritizing Cell-Type-Specific Disease Associations From Genetically Predicted Transcriptomes in Biobank Cohorts.
9 affiliations
- 1School of Biomedical Sciences, The Chinese University of Hong Kong, Shatin, Hong Kong SAR, China.
- 2Eric and Wendy Schmidt Center, Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.
- 3Department of Psychiatry, The University of Hong Kong, Hong Kong Island, Hong Kong SAR, China.
- 4KIZ-CUHK Joint Laboratory of Bioresources and Molecular Research of Common Diseases, Kunming Institute of Zoology and The Chinese University of Hong Kong, Kunming, China.
- 5Department of Psychiatry, The Chinese University of Hong Kong, Hong Kong SAR, China.
- 6CUHK Shenzhen Research Institute, Shenzhen, China.
- 7Margaret K.L. Cheung Research Centre for Management of Parkinsonism, The Chinese University of Hong Kong, Hong Kong SAR, China.
- 8Brain and Mind Institute, The Chinese University of Hong Kong, Hong Kong, Hong Kong SAR, China.
- 9Hong Kong Branch of the Chinese Academy of Sciences Center for Excellence in Animal Evolution and Genetics, The Chinese University of Hong Kong, Hong Kong, Hong Kong SAR, China.
Abstract
Integrating single-cell RNA sequencing(scRNA-seq) with the phenotypic depth of population-scale biobanks is essential for elucidating the cellular basis of complex diseases. However, this is hindered by limited scRNA-seq sample size and the lack of cellular resolution in biobanks. We present CausalCellInfer, a scalable computational framework bringing cellular resolution to bulk and genotype-imputed transcriptomes. CausalCellInfer employs an invariant causal prediction-inspired method(scRNA-seq-based invariant generalized covariance measure, scI-GCM) to select environment-stable marker genes, a parsimonious deep neural network for cell-fraction deconvolution, and regularized matrix completion to reconstruct cell-type-specific (CTS) expression profiles. This design is optimized for biobank-scale data with technical heterogeneity and limited gene overlap. Validated across simulated data, pseudo-bulk and real peripheral blood mononuclear cell (PBMC) datasets, CausalCellInfer demonstrated superior accuracy and efficiency over existing methods. Applied to ∼500 000 UK Biobank (UKBB) participants across 29 traits, it captured known pathological shifts, such as reduced pancreatic β-cell proportions in type 2 diabetes, and identified novel biological signals, including disrupted excitatory neuron-oligodendrocyte interactions in depression. Inferred CTS differential expression patterns showed significant concordance with independent single-cell studies and were enriched for Open Targets disease genes. Overall, CausalCellInfer bridges single-cell insights and population-scale genomics, enabling systematic discovery of putative disease mechanisms at cellular resolution.
Related
- Organ-Differentiated Metabolomic Profiling Reveals Putative Associations Between Predicted Bioactivity and Database-Annotated Flavor in Medicinal–Edible Talinum paniculatum (Jacq.) Gaertn
- Pan-cancer multi-omics profiling of SRPK1 identifies tumour-specific associations with prognosis and predicted chemotherapy response in breast, colon and prostate cancer
- A Drug-Referenced Machine-Learning Framework for Prioritizing Korean Herbal Formulations in Hypertension Using Linear SVM and Graph Convolutional Networks
- T cell transcriptomes in simulated microgravity in vivo
- BulkFormer: A large-scale foundation model for bulk transcriptomes
BibTeX
@article{yin2026causalcellinfer,
title = {{CausalCellInfer: Prioritizing Cell-Type-Specific Disease Associations From Genetically Predicted Transcriptomes in Biobank Cohorts.}},
author = {Liangying Yin and Yujia Shi and Ruoyu Zhang and Yong Xiang and Jinghong Qiu and Pak-Chung Sham and Hon-Cheong So},
journal = {Advanced science (Weinheim, Baden-Wurttemberg, Germany)},
year = {2026},
month = oct,
doi = {10.1002/advs.78197},
url = {https://europepmc.org/article/MED/42875726},
note = {CC BY},
}
RIS
TY - JOUR
TI - CausalCellInfer: Prioritizing Cell-Type-Specific Disease Associations From Genetically Predicted Transcriptomes in Biobank Cohorts.
AU - Liangying Yin
AU - Yujia Shi
AU - Ruoyu Zhang
AU - Yong Xiang
AU - Jinghong Qiu
AU - Pak-Chung Sham
AU - Hon-Cheong So
PY - 2026
DA - 2026/10/14
JO - Advanced science (Weinheim, Baden-Wurttemberg, Germany)
DO - 10.1002/advs.78197
UR - https://europepmc.org/article/MED/42875726
AB - Integrating single-cell RNA sequencing(scRNA-seq) with the phenotypic depth of population-scale biobanks is essential for elucidating the cellular basis of complex diseases. However, this is hindered by limited scRNA-seq sample size and the lack of cellular resolution in biobanks. We present CausalCellInfer, a scalable computational framework bringing cellular resolution to bulk and genotype-imputed transcriptomes. CausalCellInfer employs an invariant causal prediction-inspired method(scRNA-seq-based invariant generalized covariance measure, scI-GCM) to select environment-stable marker genes, a parsimonious deep neural network for cell-fraction deconvolution, and regularized matrix completion to reconstruct cell-type-specific (CTS) expression profiles. This design is optimized for biobank-scale data with technical heterogeneity and limited gene overlap. Validated across simulated data, pseudo-bulk and real peripheral blood mononuclear cell (PBMC) datasets, CausalCellInfer demonstrated superior accuracy and efficiency over existing methods. Applied to ∼500 000 UK Biobank (UKBB) participants across 29 traits, it captured known pathological shifts, such as reduced pancreatic β-cell proportions in type 2 diabetes, and identified novel biological signals, including disrupted excitatory neuron-oligodendrocyte interactions in depression. Inferred CTS differential expression patterns showed significant concordance with independent single-cell studies and were enriched for Open Targets disease genes. Overall, CausalCellInfer bridges single-cell insights and population-scale genomics, enabling systematic discovery of putative disease mechanisms at cellular resolution.
SN - 2198-3844
ER -