The sequencing microscope - a path to look at the molecules of biology

Projektinformationen

qScope

ID Finanzhilfevereinbarung: 101097367

DOI

10.3030/101097367

EK-Unterschriftsdatum 8 Juni 2023

Startdatum 1 Januar 2024

Enddatum 31 Dezember 2028

Finanziert unter

• European Research Council (ERC)

Gesamtkosten

€ 2 500 000,00

EU-Beitrag

€ 2 500 000,00

Koordiniert durch

KAROLINSKA INSTITUTET
Sweden

Periodic Reporting for period 1 - qScope (The sequencing microscope - a path to look at the molecules of biology)

Berichtszeitraum: 2024-01-01 bis 2026-06-30

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

Spatial transcriptomics — being able to read which RNA molecules sit where inside a tissue — has become a central tool for understanding how complex biological systems work, from healthy regeneration to cancer. Today's spatial-transcriptomics methods all rely, in one way or another, on a microscope: either to image the tissue directly, or to read out fluorescent or sequencing-by-imaging signals. This makes them slow, costly, two-dimensional in practice, and accessible only to laboratories with specialised optical infrastructure.

The qScope project asks a different question: what if a sequencer alone could replace the microscope? The idea is to amplify barcoded DNA copies of the molecules of interest directly inside a fixed tissue sample, let neighbouring amplicons fuse with one another locally, and then computationally reconstruct a three-dimensional spatial map of the original molecules from the resulting network of DNA-sequence neighbours. Because the technology grows in power alongside the exponential growth of sequencing capacity and computing, it has the potential to make deep, sub-cellular, three-dimensional spatial transcriptomics — and eventually spatial proteomics — broadly accessible to any laboratory with a wet-lab and access to a sequencer.

The expected impact is twofold. Scientifically, the project will deliver a generic platform for sequencing-based imaging that, in the longer term, can be used to study problems where current spatial methods fall short, including immunology and stem-cell biology. Translationally, by being substantially simpler and cheaper than microscopy-based alternatives, the technology has the potential to enable spatial molecular profiling far beyond today's well-resourced specialist laboratories.

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

During the first half of the project the team has focused on the technology-development objective foreseen for the early years of the workplan: building, testing and benchmarking the qScope chemistry and computational pipeline.

Reaction chemistry. The core in-situ amplification chemistry has been re-engineered around rolling-circle amplification (RCA) of seed and sub-seed monomers, and a one-pot protocol combining RCA, restriction digestion and PCR has been developed. Optimised RCA conditions now produce robust geometry-layer concatemers. Crosslink and template-switching artefacts have been substantially reduced through several mitigations: emulsion PCR during library preparation, mineral-oil isolation of the gel during cycling, splint-circularisation of products of interest, and a new family of NOPE-style oligonucleotides specifically designed to silence long-range crosslinks in the cover gel of qScope samples.

Ground-truth samples and reconstruction. To enable quantitative benchmarking, 3D-printed moulds have been designed that lay down arbitrary, reproducible 2D and 3D patterns of seed strands in polyacrylamide gels. These are paired with fluorescein-labelled reference gels that provide a ground-truth optical readout of the same input geometry. Using this setup, proof-of-concept reconstructions of two-dimensional (a four-target square) and three-dimensional (a three-layer sandwich) artificial samples have been obtained.

Computation. The bioinformatic pipeline has been substantially upgraded and now scales to one to two billion sequencing reads per run, the regime required for sub-cellular reconstruction of larger 3D tissue volumes. A proof-of-concept reconstruction algorithm has been implemented as a benchmark for further algorithm development.

Toward biological applications. In parallel with the technology work, preparatory activities for the project's biological applications have begun: mouse spleen tissue has been obtained, antibody targets have been redesigned, and antibody screening against potential germinal-centre targets together with antibody–oligo conjugation is underway. An additional proteomics layer based on DNA–antibody conjugates is being integrated into the platform.

Knowledge transfer. The team visited the laboratory of one of the parallel originators of the DNA-microscopy concept early in the reporting period, to align on reconstruction algorithms, share experimental pitfalls and benchmarks across groups.

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 first half of the project has delivered several advances beyond the state of the art in DNA microscopy and sequencing-based imaging.

First, the project has produced what is, to our knowledge, the first systematic demonstration of DNA-microscopy reconstructions of designed 2D and 3D ground-truth targets, with a fluorescent optical reference for direct comparison. This moves the field from a self-validating regime, in which reconstructions can only be checked against themselves, to one in which they can be benchmarked against an external truth.

Second, an RCA-based, one-pot wet-lab chemistry that combines amplification, restriction digestion and PCR in a single tube simplifies the protocol substantially relative to earlier PCR-only DNA-microscopy implementations, and reduces crosslink artefacts. New NOPE-style oligonucleotides specifically targeting cover-gel crosslinks address a problem that only became visible as the protocol matured.

Third, the two-layer architecture (a geometry layer and an independently tunable measurement layer) provides a resolution-versus-depth trade-off that earlier DNA-microscopy methods could not offer.

Fourth, the computational pipeline scales to one to two billion reads per run and is now ready for the larger sequencing budgets required for tissue-scale 3D reconstruction.

To support uptake and translation, a method patent has been filed and a spin-off company has been incorporated to take ownership of the patent and to pursue commercial development of the technology in parallel with continued academic research. Further uptake will require continued algorithm development, validation in biologically relevant tissue, and engagement with the spatial-omics community; the patent and spin-off provide a clear route for IPR-supported commercialisation.