Structure-enhanced ML methods for spatial transcriptomics cell type deconvolution
Official title: DMS/NIGMS 1:Structure-Enhanced Statistical and Machine Learning Methods for Spatial Transcriptomics Cell Type Deconvolution and its Applications
Funded at Case Western Reserve University · NIH-11535833 · Data updated 2026-11-12
A bidirectional, smarter way to figure out which cell types are present in tissue pictures from spatial gene expression data, helping brain tumor patients and researchers.
Quick facts
| Grant type | R01 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Case Western Reserve University NIH-funded |
| Lab location | 1 site (Cleveland, United States) |
| Project start | August 1, 2026 |
| Project end | May 31, 2029 |
| Funding year | 2026 |
| Award (this year) | $200,001 |
| NIH project number | 1R01GM169393-01 |
| Project ID | NIH-11535833 on NIH RePORTER |
Who could benefit from this research
✓Could help
Ideal candidates are individuals with brain tumors contributing data or samples for spatial transcriptomics analyses and researchers who will use the methods.
✕Less likely to help
Patients not involved in spatial transcriptomics studies or who are not receiving related diagnostic or therapeutic investigations may not directly benefit.
What this research studies
This project develops new machine learning methods to estimate which cell types are present in spot-based spatial transcriptomics data, making it cheaper and faster to study tissue architecture. It combines Bayesian modeling, graph-based features from tissue images, and deep learning to quantify uncertainty and tune parameters automatically. The work aims to improve cell type clustering and deconvolution, with applications to brain tumor samples provided by clinical collaborators. Researchers will test the approach on real patient data to reveal how immune cells interact with tumors in space. The goal is to produce robust tools that could be used by other scientists studying tissue microenvironments.
Why it matters
Potential benefit: If successful, patients could benefit from faster, cheaper spatial analyses that help scientists better understand tumor biology and potentially guide future therapies.
How similar studies have gone: Preliminary simulations and published-data analyses suggest promising performance for similar deconvolution and Bayesian approaches, but the exact method is novel.
Where this research is happening
-
Case Western Reserve University
Cleveland, United States Active
Researchers
- Principal investigator: Zhang, Liangliang — Case Western Reserve University
- Study coordinator: Zhang, Liangliang
Related research
Other NIH-funded projects on the same topics.
-
Estrogen signaling and how the intestinal lining protects against injury
A study at Case Western Reserve University looks at how estrogen receptors in the gut lining help protect against intestinal injury and inflammatory bowel disease.
-
Diabetes: Long-term effects of intensive glucose control on complications
A long-term study follows adults with type 1 diabetes to see how keeping blood sugar close to normal affects eye, kidney, nerve, heart, and quality of life outcomes over many years.
-
Epidemiology of Diabetes Interventions and Complications (EDIC) Study follow-up
This long-term study follows adults with type 1 diabetes to see how early intensive glucose control affects complications and quality of life over many years.
-
Role of THAP1 in brain cell support cells shaping brain development
Aims to understand how the THAP1 protein in oligodendrocyte lineage cells helps shape the brain’s environment and myelination during development, which may relate to a movement disorder called DYT6 dystonia.
-
New human monoclonal antibodies to stop malaria caused by Plasmodium vivax
A researcher team is developing human monoclonal antibodies to prevent and treat malaria from Plasmodium vivax, with a plan to test promising antibodies in primates and prepare for human trials.
-
PopCASE: A user-friendly tool to link cancer data for communities and researchers
Aimed at improving population cancer research, PopCASE helps connect patient cancer data with community information to inform better care and outreach.
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page.