Towards Perfect Molecular Measurement

Abstract

The rapid pace of technological innovation has profoundly transformed life sciences research. Whereas less than 100 years ago researchers were often confined to descriptive and theoretical practices, modern scientists can perceive the inner workings of living systems using advanced measurement technology. This ability to acquire measurements of molecules enables the construction of mechanistic models of biological phenomena. Yet despite the power of existing measurement technologies, many aspects of life remain obscured. The things which we cannot observe or measure inspire continuing technology development efforts, such as the work presented in this dissertation. Here, we present a historical, empirical, and theoretical analysis of biology to determine from first principles the shape of future technologies. As a step in this progression, we developed fluorescent in situ sequencing (FISSEQ) of RNA molecules—enabling detection of the spatial organization of gene expression with single-molecule, single-nucleotide resolution. We demonstrate that RNA FISSEQ can be used on a wide array of samples for the investigation of cellular phenotype, gene regulation, and environment in situ, and examine RNA expression and localization in human primary fibroblasts with a simulated wound-healing assay. We describe improvements of the FISSEQ library construction method, such as for targeting certain species of RNA, for sequencing DNA in situ, and for detection of proteins and other biomolecules, and integration of FISSEQ with expansion microscopy (ExM) for resolution beyond the diffraction limit of light and enhanced sensitivity. We present two novel nucleic acid sequencing chemistries, a fully-automated FISSEQ robot, and our work building advanced software for processing and analyzing FISSEQ data. We describe our ongoing use of FISSEQ in the study of cancer, development, metabolism, and neuroscience. Finally, we speculate about massively multiplex molecular detection in vivo.