Embryonic development involves the generation of the vast cellular diversity observed in an adult organism. This process requires spatially and temporally controlled cell state transitions through which a stem cell becomes a diffeentiated cell type. Thus, delineating the mechanisms that control changes in cell identity is fundamental to our understanding of organismal development. Further, since stemness can also be reactivated during oncogenesis, determining its regulation also has broad implications for cancer biology. Our lab employs the neural crest as a model system to examine the molecular control of stem cell identity in development and disease. The neural crest is a vertebrate-specific embryonic stem cell population that forms a wide range of derivatives, including neurons, chondrocytes, and melanocytes. These cells delaminate from the central nervous system and migrate through well-defined pathways within the developing embryo. Through the course of migration, they progressively lose their stem cell properties and become committed to a specific fate. Intriguingly, during tumorigenesis, neural crest stem cell identity can reemerge in adult cells, giving rise to cancers like melanoma and neuroblastoma. In the first part of my thesis research, I combined genomic analyses with classical embryology to explore the regulation of stem cell identity during avian neural crest development. This work demonstrated how Yamanaka factors, including Sox2, Oct4, and Lin28a, control neural crest multipotency by regulating the epigenome and transcriptome of this specialized stem cell population. My findings revealed that the general pluripotency program cooperates with tissue-specific factors to modulate the gene regulatory network that endows neural crest cells with their unique features. In the second part of my dissertation, I aimed to delineate the mechanisms that reactivate stem cell fate in neural crest-derived cancers such as melanoma. By employing genomic and functional analysis, I discovered that neural crest cells share many molecular features with cancer cells, including the metabolic adaptation Warburg effect and reliance on the same signal transduction pathway for migration. Further, by profiling melanocytic progenitors and human melanoma cells, I identified thousands of neural crest-specific epigenomic regions that reemerge during malignant transformation to establish a stem-like state within the tumor. Together, these results highlight how developmental regulatory networks are co-opted during cancer progression to promote tumor heterogeneity