ABSTRACT The overall goal of the proposed research program is to improve our understanding of single cell biology through information maximization techniques, by applying molecular engineering and computational approaches in sequencing. Specifically, single cell sequencing is rapidly becoming the predominant method for studying human biology and disease because it removes the confounding factor of sequencing cell mixtures in bulk. However, it has major pitfalls: significant material consumption during library preparation, noisy data readouts and signal dropout, and unclear paths for data integration across datasets. The overall vision of the proposed research program is to develop a pan-omic analysis strategy that enables perpetual re-use of any single cell source material. It revolves around a hybrid molecular engineering and computational framework that is loosely inspired by principles found in computing. The experimental core of the proposed research program revolved around a new molecular technology referred to as APEX (‘Attachment- based Primer EXtension’). The major innovation of APEX is the covalent conjugation of genomic material (i.e. DNA or cDNA) to a solid phase support such as an agarose magnetic bead, followed by utilizing only polymerase- based assays for non-destructive molecular interrogation. In this project, we will focus APEX development on single cell transcriptome sequencing applications, with general applicability to genome biology. As a model system, we will utilize peripheral lymphocytes as they consist of complex subpopulations with distinct characteristics at multiple levels of omic features. The project will focus on assay development and optimization, development of bioinformatic algorithms for data integration, and scale up to large cohorts as a demonstration of the scalability of the technology.

NARRATIVE Human samples contain immense riches of data, but are extremely precious as experiments consume them. To solve this problem, we propose developing a new method that indefinitely preserves such material across experiments. This means ideally limitless experiments can be performed to improve accuracy, or different ones can be performed to look at different types of information afforded by the sample.