Variability in gene regulation is a

Variability in gene regulation is a fundamental characteristic of biology, allowing cellular adaptation in many states, such as development, stress response, and survival. In early disease onset, genetic and epigenetic variability permit the formation of multiple cellular phenotypes. In cancer, increased cellular plasticity ultimately results in the foundation of a tumor with the phenotypic alterations necessary to dynamically adapt, proliferate, metastasize, and acquire therapeutic resistance throughout the course of the disease. One prominent form of cellular regulation is DNA methylation, an epigenetic chemical modification that can alter gene expression. Hypermethylation-induced silencing is known to occur early on in tumorigenesis, often in precursor phases of the disease. Furthermore, tumors have been shown to undergo epigenetic reprogramming throughout progression of the disease. In light of these observations, methylation heterogeneity may serve as a novel biomarker for early cancer detection. Early detection of cancer remains challenging, as symptoms often manifest in later stages and current screening techniques often lack the requisite sensitivity and specificity. To maximize effectiveness, routine screening techniques should be noninvasive, simple, and unbiased. To this end, liquid biopsies (e.g. blood samples) containing cellular debris, such as tumor-derived cell-free DNA in the plasma, are ideally suited towards routine screening. However, detection of tumor-derived molecules in plasma is challenging, as they are often rare and may be eclipsed by a high background of molecules from healthy cells. Thus a sensitive platform capable of quantifying epigenetic heterogeneity could uncover new insights and improve early detection. In this dissertation, I present a microfluidic digital melt platform for facile, highly-sensitive detection and molecule-by-molecule profiling. The platform is applied towards the quantification of epiallelic heterogeneity. Digitization of rare molecules into thousands of microchambers followed by parallelized sequencing interrogation through high resolution melt enables order of magnitude higher sensitivity than current techniques and insight into new intermolecular characteristics. I also demonstrate how this platform may be modified to complement and improve the sensing capabilities of existing commercial technologies. Finally, I validate the potential clinical utility of this platform through detection of methylation heterogeneity in complex clinical samples towards noninvasive screening applications. The technical capabilities along with the operational simplicity of this platform facilitate adoption by other laboratories and offer potential clinical utility. This system may offer new insights into the mechanisms of epigenetic regulation in pathogenesis, and potentially improve early diagnosis.