Mode
Text Size
Log in / Sign up

Multi-omics integration identifies candidate genes and cell-type-specific effects in glioma.

Multi-omics integration identifies candidate genes and cell-type-specific effects in glioma.
Photo by Google DeepMind / Unsplash
Key Takeaway
Note that multi-omics integration identifies candidate genes, but single-cell mechanisms remain poorly understood.

This computational analysis integrated genome-wide association study (GWAS), bulk tissue, and single-cell multi-omics data to investigate glioma and glioblastoma. The study aimed to prioritize genetically supported candidate genes and identify cell-type-specific causal effects. No specific patient population size or clinical setting was reported for this computational integration.

The analysis prioritized 11 high-confidence and 47 putatively causal genes. Trait-relevant populations were identified as astrocytes and oligodendrocyte precursor cells (OPCs). Communication between astrocytes/OPCs and neurons was found to be significantly increased. Additionally, 14 cell-type-specific causal effects were discovered.

High-confidence causal genes included EGFR in astrocytes, CDKN2A in OPCs, and JAK1 in excitatory neurons. Twelve out of 14 effects, representing 85.7%, were associated with non-glioblastoma-relevant cells. Secondary outcomes included target enrichment, druggability, genetic pleiotropy, and drug repurposing potential.

Safety and tolerability data were not reported, as this was a computational study without patient intervention. A key limitation is that single-cell-level mechanisms underlying gliomagenesis are poorly understood. While the practice relevance involves advancing targeted precision therapeutics, the distinction between high-confidence and putatively causal genes must be maintained. Causality remains investigational given the current understanding of these mechanisms.

Study Details

EvidenceLevel 5
PublishedMar 2026
View Original Abstract ↓
BackgroundGliomas constitute 80% of malignant brain tumors, with glioblastoma (GBM) being the most aggressive subtype. The single-cell-level mechanisms underlying gliomagenesis are poorly understood, hindering therapeutic development. We combine genome-wide association studies (GWAS) with bulk tissue and single-cell multi-omics to prioritize gliomagenesis genetically supported candidate genes and reveal cell-type-specific biological mechanisms. MethodsWe integrated the largest glioma GWAS with brain-specific multi-omics to prioritize genetically supported candidate genes using two broad categories of prioritized methods. Biological enrichment, differential gene expression, and CRISPR/miRNA were used to assess target enrichment and druggability. By integrating single-cell multi-omics data (genomics, transcriptomics, epigenomics), we investigated GBM-relevant cells, tumor microenvironment (TME) interactions, and cell-type-specific mechanisms in glioblastomagenesis. Additionally, phenome-wide association studies (PheWAS) and drug repurposing analyses were conducted to annotate genetic pleiotropy and enhance drug repositioning. ResultsWe prioritized 11 high-confidence and 47 putatively causal genes, most of which are druggable. Astrocytes and oligodendrocyte precursor cells (OPCs) were identified as the trait-relevant populations in GBM, with significantly increased TME cell communication between these populations and neurons. Fourteen cell-type-specific causal effects in glioblastomagenesis were discovered, including three high-confidence genes (EGFR in astrocytes, CDKN2A in OPCs, and JAK1 in excitatory neurons). Most effects (85.7%, 12/14) were associated with non-GBM-relevant cell cells, encompassing both glial and neural cells. ConclusionsThis study systematically identifies genetically supported candidate genes in gliomagenesis and their cell-type-specific effects, providing insights into the cell-resolved mechanisms of glioma susceptibility and advancing targeted precision therapeutics.
Free Newsletter

Clinical research that matters. Delivered to your inbox.

Join thousands of clinicians and researchers. No spam, unsubscribe anytime.