Spatial transcriptomics chips with sequencing-based microscopy

Projektinformationen

MESH_CHIP

ID Finanzhilfevereinbarung: 101138356

DOI

10.3030/101138356

Projekt abgeschlossen

EK-Unterschriftsdatum 31 August 2023

Startdatum 1 Januar 2024

Enddatum 30 Juni 2025

Finanziert unter

• European Research Council (ERC)

Gesamtkosten

Keine Daten

EU-Beitrag

€ 150 000,00

Koordiniert durch

KUNGLIGA TEKNISKA HOEGSKOLAN
Sweden

Periodic Reporting for period 1 - MESH_CHIP (Spatial transcriptomics chips with sequencing-based microscopy)

Berichtszeitraum: 2024-01-01 bis 2025-06-30

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

Spatial transcriptomics is a widely used and transformative approach that has been used to reveal how gene expression varies across tissues, facilitating insights in development, disease, and cell biology. However, current technologies rely on costly and low-throughput manufacturing of barcoded surfaces for capturing genetic material. This is done today with lithography or microarrayp rinting techniques (e.g. 10x Visium) or labor-intensive in situ decoding of randomly scattered arrays. These limitations restrict spatial transcriptomics to well-funded research applications, leaving high-impact diagnostics, large-scale mapping, and routine clinical deployment largely unaccessible.

The MESH CHIP project introduces a paradigm shifting-solution to this bottleneck. Rather than relying on traditional spatial fabrication, MESH CHIPs are constructed through a self-assembling DNA process that forms dense networks of spatially informative capture regions on surfaces. Importantly, spatial information is not pre-encoded but rather reconstructed afterward from sequencing data using graph-theoretic and machine learning methods. This "posterior localization" strategy avoids the bottlenecks of physical localization and enables a highly scalable, cost-effective, and resolution-flexible approach to spatial transcriptomics.

The project aims to translate this breakthrough concept into a commercially viable technology by refining both the biochemical fabrication and computational reconstruction pipeline, benchmarking against existing state-of-the-art platforms, and validfating performance in a biological model. The expected impact is that the MESH CHIP could unlock spatial transcriptomics for high throughput research and biopsy-scale diagnostics. This is in line with the trend of decentralizing genomic workflows and further democratization of access to omics data worldwide.

By shifting spatial resolution from expensive hardware to scalable computation, MESH CHIP reframes spatial transcriptomics as a software-solvable problem. Its success would improve access to powerful genetic mapping and has the potential to reshape how future genomics tools are conceptualized - favoring molecular self assembly approaches over traditional precision engineering.

Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse

Over the course of the MESH CHIP project, we made progress in establishing and validating the core technical elements required for a new, self-assembly and computation approach to spatial transcriptomics. Our activites focused on refining surface fabrication, improving the fidelity of DNA barcode network formation, and developing computational tools for spatial reconstruction from sequencing data. We refined protocols for polony generation on functionalized surfaces and developed a dedicated apparatus that enables reproducible surface fabrication. We improved the biochemical steps required for DNA interlinkage, leading to higher yield of network connections. An important insight that emerged during development was the identification of false or noisy barcode connections as a source of distortion, leading us to develop computational tools to detect such noise. This led to a paper "Spatial coherence of DNA barcode networks" (Bonet et al, 2024). These technical achievements have advanced the technological readiness level (TRL) from early stage laboratory validation (3-4) to prototype demonstration (TRL 5). The ability to reproducibly fabricate dense, spatially informative DNA surfaces and computational reconstruct spatial structure from sequencing data alone is a major de-risking step towards a minimum viable product.

Outcomes of the Action

1. Developed a robust and reproducible protocol for forming dense, high-quality polony surfaces.
2. Improved interlinking chemistry for generating spatially reconstructable barcode networks.
3. Successfully reconstructed spatial structure from sequencing data alone, validating the principle of posterior localization.
4. Produced a computational framework (STRND) and a theoretical contribution on spatial coherence, submitted to a high-impact journal.
5. Clarified technical bottlenecks and de-risked foundational steps, setting up for future tissue-level experiments.

Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)

The MESH CHIP project has laid a technical and conceptual foundation for a new class of spatial transcriptomics technologies that shift spatial resolution from hardware into the domain of computation. The project demonstrated key milestones such as reproducible generation of dense surface polonies, robust interlinkage chemistry for encoding spatial proximity, and reconstruction of spatial organization using only sequencing data using a computational strategy with the potential to disrupt the current industry paradigm.

The project resulted in several key technical and scientific outcomes. We successfully refined the protocols for polony generation and interlinking on functionalized surfaces, establishing a reproducible method for creating spatially informative DNA barcode networks. Using synthetic data, we demonstrated proof-of-concept spatial reconstructions with our STRND algorithm, validating the core concept of posterior localization. In parallel, we developed and submitted for revision a scientific manuscript in which we introduced spatial coherence metrics to assess the geometric integrity of DNA barcode networks, contributing foundational tools for sequencing-based microscopy. On the intellectual property front, we filed and obtained a granted Swedish patent covering the core method for generating networked 2D polony surfaces.

The potential impact of the results of this project is that we have made both technical-side and exploitation-side progress towards a viable commercial product, the MESH CHIP, that could unlock broad adoption of spatial transcriptomics in settings previously out of reach due to cost and scalability barriers. This includes high-throughput basic research (e.g. brain mapping, pharmaceutical tissue screening, clinical biopsies, and eventually routine diagnostics. By radically reducing cost per surface and eliminating the need for complex fabrication or imaging equipment, MESH CHIP could present a potential path toward spatial transcriptomics as a scalable, modular, and easily consumable solution to a wide space of domains.

The needs for further uptake and success span both technological development and commercial exploitation. On the technical side, the priority is to validate transcriptomic capture performance using biological tissue samples. In parallel, comparative benchmarking against leading spatial transcriptomics technologies like 10x Visium, Curio, etc will be necessary to quantify advantages. From an IP perspective, while a Swedish patent has been granted and a PCT application is under review, continued support is needed to navigate examiner feedback and pursue national phase entry. On the commercialization side, a startup company is being founded by the team. Pre-seed funding has been secured through KTH Innovation and KTH Holding AB, and the team is receiving guidance from a dedicated business coach. Establishing partnerships with early adopters in pharma, diagnostics, and academia will be important for validating use cases and guiding product development.