Cancer is believed to be driven by genetic mutations. However, mouse models and human patients with cancer-predisposing germ line mutations show that only a small fraction of their cells become malignant, suggesting mutations themselves are not sufficient to cause transformation and additional determinants must exist. There are two prevailing, non-exclusive models to account for the additional determinants. One is that full transformation requires accumulation of multiple genetic mutations, which posits transformation occurs only when all mutations have been acquired. The recent success in reprogramming malignant cells into pluripotent stem cells indicates that normal cells of essentially all lineages can exist in the presence of replete cancer mutations. The second model postulates that mutations in rare stem cells initiate malignancy. Advancement in sequencing technology has led to the detection of cancer mutations in the hematopoietic stem cells (HSCs) of healthy individuals. The prevalence of mutation-bearing preleukemic HSCs questions whether stem cells are the direct target of transformation. These conflicts demand a new conceptual framework for how malignancy arises from the cell-of-origin. To reconcile with the new insights brought about by new technologies, we propose to define the molecular and cellular nature of cancer cell-of-origin to identify the missing determinant for malignant transformation. We would like to capture and visualize the entire malignant transformation process from a normal blood progenitor transitioning into a cancer cell, and examine the unique features of these rare cells founding cancer.