Researchers at the Max Delbrück Center have released openDVP, an open-source software platform that simplifies the spatial mapping of proteins inside tissues. Published in Nature Methods, the tool combines artificial intelligence and mass spectrometry to help scientists analyze cellular neighborhoods in diseases like lung and breast cancer.
Satellites map Earth’s geography down to the individual tree, but tracking proteins inside human tissue has long remained a fragmented puzzle. Dr. Fabian Coscia and his team at the Max Delbrück Center introduced openDVP to solve this bottleneck. By streamlining workflows that once required specialized computational expertise, the platform opens high-resolution spatial proteomics to a global research community.
Key Terms for Spatial Proteomics and Mass Spectrometry
- Spatial Proteomics: A laboratory technique that measures the quantity and physical location of proteins inside specific cells within a tissue sample.
- Mass Spectrometry: An analytical method used to identify molecules, allowing researchers to catalog thousands of proteins simultaneously.
- Cellular Neighborhoods: Functional groups of specialized cells that cluster together to perform coordinated tasks within the body.
Accelerating Proteomic Mapping With Open-Source Software
Dr. Fabian Coscia’s Spatial Proteomics group originally co-developed Deep Visual Proteomics, known as DVP, which debuted in scientific literature in 2022. The method integrates high-resolution imaging, artificial intelligence, and mass spectrometry.
To overcome this hurdle, the team engineered openDVP. The software refines the image analysis pipeline, allowing researchers to isolate and measure specific cells or tissue zones without needing advanced programming skills. Accompanying this software is a new sampling strategy named cellular neighborhood-guided DVP, or cnDVP. First author Jose Nimo noted that cnDVP accelerates tissue sampling by roughly 100 times compared to earlier workflows.
| Tissue Analyzed | Key Biological Target Identified | Clinical Relevance |
|---|---|---|
| Placenta | Cellular communication networks | Illuminates details of cellular communication. |
| Lungenadenokarzinom | GLUT1 glucose transporter | Linked to glucose uptake in metastatic cells and reduced patient survival rates. |
| Breast Tumor | Inflammatory signaling pathways | Associated with resistances against chemotherapy. |
Uncovering Therapeutic Vulnerabilities in Oncology
To test the open-source pipeline, the research group processed three distinct human tissue types: placenta, lung adenocarcinoma, and breast cancer. In the lung cancer samples, the software isolated the GLUT1 glucose transporter, a protein that helps metastatic tumor cells with glucose uptake. Patients exhibiting high levels of GLUT1 demonstrate worse survival rates, highlighting its clinical relevance.
In breast cancer tissue, the technology surfaced increased inflammatory pathways connected to chemotherapy resistance. By mapping these microenvironments, the platform helps scientists pinpoint why certain malignancies stop responding to medical therapies. The open-source architecture ensures that laboratories worldwide can apply identical pipelines to diverse pathology samples.
Researchers Aim to Integrate AI Foundation Models
Looking ahead, the research group aims to integrate large AI foundation models trained on histopathological data. This computational expansion will allow teams to conduct large-scale spatial proteomic studies across many different diseases.
References
- Max-Delbrück-Centrum für Molekulare Medizin (MDC): Research Announcement on Spatial Proteomics
