Sjögren's disease & overlap diseases
Uncovering CD248, MMP28, and SLC16A10 in primary Sjögren's disease: a machine learning-driven SHAP approach for CD4+ T cell-associated biomarker discovery
Authors' conclusion
Our bioinformatic study identifies CD248, MMP28 and SLC16A10 as candidate biomarkers and therapeutic targets for SjD, with their dysregulation specifically enriched in CD4+ T cell subsets, unveiling a previously underappreciated mechanism in SjD pathogenesis.
Study context
- Population
- Transcriptomic datasets of sjögren's disease (SjD) patients and healthy controls
- Participants
- 382 samples
- Comparison
- SjD patients vs healthy controls
- SjDn = 321
- Healthy controlsn = 61reference
- Outcome
- Diagnostic signature (hub) genes identified by machine learning and SHAP
Methodological detailsbioinformatic, 12 machine learning algorithms, SHAP interpretable modelling, external dataset validation, single-cell RNA sequencing (scRNA-seq), molecular docking
- Bioinformatic
- 12 machine learning algorithms
- SHAP interpretable modelling
- External dataset validation
- scRNA-seq
- Molecular docking
Main finding
As reported in the abstract
Hub genes identified (CD248, MMP28, SLC16A10)
- Significant differential expression in external datasets
- p<0.05
- Diagnostic performance in external datasets
- AUC >0.7
Further findings
Immune infiltration
- Positive correlation of CD248/SLC16A10 with naive CD4+ T cells
- P < 0.05
- Positive correlation of SLC16A10/MMP28 with memory resting CD4+ T cells
- P < 0.05
Single-cell localisation
- CD248 predominantly in naive CD4+ T cells; SLC16A10 and MMP28 in both naive and memory CD4+ T cell subsets
Molecular docking
- Stable targeting of CD248, MMP28 and SLC16A10 by AZA, LEF, MTX, HCQ, iguratimod, pilocarpine and cevimeline
Original publication
Uncovering CD248, MMP28, and SLC16A10 in primary Sjögren's disease: a machine learning-driven SHAP approach for CD4+ T cell-associated biomarker discovery. Clin Exp Rheumatol. Published 6 June 2025. doi:10. 55563/ clinexprheumatol/ 2ffdc4
Composed by AI from the published abstract and checked against it. How Briefs are made →