From Maps to Medicine

In late 2000s, towards the end of my graduate school, the field of “System Biology” started to emerge. A good part of my work during my postdoc and later when I started in industry, focused on characterizing biological phenotype using multimodal data1,2. These experiences also helped me appreciate and imagine what would be ideal. If only there was a way to understand the “system” with a spatial and temporal resolution. Clearly that was beyond the realm of possible at the time.

Fast forward to 2024, a part of that wishlist has become a reality. With years of technology development Spatial Biology has reached an inflection point. A number of recent publications, some highlighted below, strongly suggest that the technology has matured enough to provide actionable insights, advancing our fundamental understanding of biology as well as applications in the clinic. Recently, Nature Methods named Spatial Proteomics (Nature Methods, 2024) as Method of the Year, underscoring how spatial biology is bridging the gap between foundational discoveries and clinical applications. This recognition further validates the transformative potential of spatial technologies and strengthens my conviction in their ability to reshape medicine.

This image is made by Georg Walmann at Max Planck Institute of Biochemistry. The “shoreline” is the epidermis and distinct cellular types within the dermis inner layer are tagged, involved with the life-threatening condition of toxic epidermal necrolysis. Adapted from: The Dawn of Spatial Medicine

Breakthroughs Start with Curiosity but Demand Impact

Spatial Biology is redefining how we interpret and apply this knowledge. The real breakthroughs come not from discovery alone but from turning that discovery into actionable insights that fundamentally reshape our understanding of biology.

Spatial Transcriptomics:  

Mapping RNA activity in context provides a deeper understanding of how gene expression is influenced by the spatial organization of cells within tissues. Jain et al., used spatial transcriptomics at single-cell resolution to map functional tissue units (FTUs) in human organs like the kidney and intestine. By integrating this with multi-omics methods, they uncovered cellular diversity, tissue-specific microenvironments, and key gene expression patterns, such as ligand-receptor signaling dynamics. This approach offers unprecedented insights into tissue organization, function, and disease mechanisms, while enabling cross-organ comparisons. Spatial transcriptomics links gene expression to its spatial context, providing a deeper understanding of tumor microenvironments, neuronal communication, and other complex systems, paving the way for more precise therapeutic strategies.

Spatial Proteomics:

One story that strongly reflected the 360 degree impact of such analysis is the TEN study. Using Deep Visual Proteomics, researchers didn’t just reveal the activation of the JAK/STAT pathway—it provided a path to treatment that saved lives within days. JAK inhibitors, tested first in preclinical models, were administered to seven TEN patients. Within just 48 hours, a striking recovery began—something unprecedented for a condition with no prior effective treatment. All seven patients fully recovered, with no adverse effects reported. That’s the kind of transformation we’re here for. DVP This is the kind of leap that reminds us why we entered science in the first place: to create tools that can truly impact the world.

Spatial proteomics is also being leveraged to explore cancer progression, autoimmune diseases, and metabolic disorders, offering insights that drive both drug discovery and diagnostic advancements. By integrating protein data with other omics layers, it paints a comprehensive picture of cellular functionality in health and disease.

Spatial Metabolomics

When Peruzzi et.al identified metabolite shifts in gliomas using MALDI-based metabolomics, it wasn’t just a technical achievement. It was a glimpse into the future of personalized medicine—where we don’t just see what’s happening but understand why. Using MALDI-based tumor metabolomics, they identified changes in the glutathione detox pathway in response to lapatinib exposure, aligning with prior evidence of EGFR inhibition altering glutathione metabolism. This approach not only confirmed drug-specific effects but also allowed the discovery of metabolite signatures associated with therapeutic responses. By integrating metabolomics data with spatially resolved drug exposure information, the study provided insights into localized metabolic shifts and their role in tumor pharmacodynamics.

Combining spatial metabolomics with RNA and protein data opens the door to a holistic understanding of phenotypes, offering clues about not just what is happening in a cell but why. This integration is the key to developing targeted therapies and monitoring their impact in real time.

Integration is The Only Way to Decipher Disease Complexity

The future of spatial biology is nothing short of pivotal. But let’s be honest: the field is at a critical juncture. We’ve moved past the era of generating data for the sake of discovery; now it’s about making that data work for us. The biggest roadblock? We’re drowning in complexity without the right systems to extract actionable insights.

It’s time to stop tinkering around the edges and address the foundational challenges head-on. The volume and complexity of spatial datasets demand a paradigm shift in how we process, harmonize, and analyze them. Anything less than robust, scalable solutions is simply inadequate.

Here’s the uncomfortable truth: we’re generating incredible spatial data, but our ability to integrate it across RNA, protein, and metabolite layers is still lagging. That’s where my conviction comes in. This is solvable—but not without robust data infrastructure, AI-driven harmonization, and scalable computing. Without these, spatial medicine will remain a niche field instead of the mainstream tool it’s destined to become.

Without the Right Data Infra and AI, Scalable Integration Will Remain Out of Reach!

Here’s what needs to happen—no compromises, no half-measures:

1. Scalable Data Ingestion and Processing: The volume and complexity of spatial datasets necessitate pipelines capable of automated ingestion, harmonization, and quality control at scale. AI can standardize data from disparate sources, ensuring uniformity without compromising biological nuance.
2. Data Harmonization and Integration: Beyond format alignment, harmonization involves contextualizing data to account for cellular, spatial, and temporal relationships. Ontologies and semantic frameworks play a crucial role here, enabling structured integration across omics layers.
3. AI for Feature Extraction and Pattern Recognition: Machine learning models excel in identifying patterns across high-dimensional datasets, revealing previously hidden relationships. By integrating RNA, protein, and metabolite data, these models can generate predictive insights about phenotype and therapeutic responses.
4. Cost-Effective Computational Infrastructure: Spatial omics demands advanced cloud-based or high-performance computing systems to process large datasets efficiently. Investment in such infrastructure ensures accessibility and scalability, crucial for democratizing spatial biology research.

Solving these challenges isn’t just about enabling spatial biology—it’s about creating a framework for the next generation of biological discovery, where insights are integrated, scalable, and actionable at unprecedented levels. The time to build this foundation is now.

Spatial Biology will drive enormous opportunities in patient care in the next 3-5 years!

In the next 3-5 years, Spatial Biology will become a mainstream tool in the repertoire of R&D teams. Think scRNASeq. We saw a broad scale adoption of this technology in the last 5 years. Just to add more context, it takes 3-5 years on an average to go from a target identification to IND. So if you are getting started on your IND journey you might want to seriously consider how Spatial Biology can help. If you are thinking of indication expansion, you might want to make Spatial Biology part of your plans.

The challenges are significant,but the opportunities are enormous. This isn’t just a moment in science—it’s a movement. And I, for one, am all in.

Let’s chart this map together and hope that the next frontier of temporal resolution is within the realm of possibility in the next wave of breakthrough!

-Abhishek Jha | Co-Founder & CEO | Elucidata