In situ DNA sequencing-based microscopy for subcellular spatial transcriptomics

ongoing

New tools are needed in spatial transcriptomics, which uses imaging to resolve the positions of RNA in their native biological contexts to reveal molecular mechanisms underlying cell states and interactions. The developing embryo exhibits complex transcriptional regulation during the maternal-to-zygotic transition, when the reservoir of maternal RNA is phased out and the zygotic genes are turned on. Existing spatial sequencing approaches cannot simultaneously achieve subcellular resolution, whole transcriptome coverage, isotropic 3D resolution, and the parallel mapping of proteins and genetic regulatory elements that is desirable to form a mechanistic picture of transcriptional regulation in any system as complex is the embryo. The nascent field of DNA microscopy is perfectly suited for the high-throughput, multiplexed, molecular mapping needs of such problems. DNA microscopy uses carefully engineered in situ PCR, next-generation sequencing, and the mathematics of stochastic geometry instead of optics to convey microscale spatial information. In 2018, I developed a 2D DNA microscopy approach based on topological reconstruction of adjacent patches of barcoded DNA “polonies”. My work is among a few papers appearing just in the last year that together constitute a new field. I will adapt my 2D topological DNA microscopy method to one based on in situ whole-transcriptome sequencing in 3D, aiming to achieve subcellular resolution. I will also develop the mathematical basis of topological reconstruction and new computational tools to deal with the 3D data. Finally, I aim to incorporate the capability to localize other molecules in parallel with the transcriptome such as oligonucleotide-conjugated antibodies targeting specific transcription factors and other genetic regulatory elements. The technique, deployed on developing C. elegans embryos, will be used to study spatially dependent regulatory mechanisms such as the determination of cell polarity and fate during cleavage in greater breadth and depth than ever previously achieved. By being optics-free, this has the potential to overcome fundamental limitations imposed by traditional forms of microscopy, greatly expand the ease and throughput of spatial transcriptomics, and pave the way for routine use of hyper-multiplexed molecular imaging.