Cancer cell migration is essential for the early steps of metastasis, during which cancer cells move through the primary tumor and reach the blood vessels. In vivo, cancer cells are exposed to directional guidance cues, either soluble, such as gradients of growth factors, or insoluble, such as collagen fiber alignment. Depending on the number and strength of such cues, cells will migrate in a random or directed manner. Interestingly, similar cues also stimulate cell proliferation. In this regard, it is not clear whether cell cycle progression affects migration of cancer cells and whether this effect is different in random versus directed migration. In this study, we tested the effect of cell cycle progression on random and directed migration, both in 2D and 3D environments, in the breast carcinoma cell line, FUCCI-MDA-MB-231, using computational image analysis by LEVER. Directed migration in 2D was modeled as chemotaxis along a gradient of soluble EGF inside 10 µm-wide microchannels. In 3D, directed migration was modeled as contact guidance (alignotaxis) along aligned collagen fibers. Time-lapse recordings of cells in 2D and 3D revealed that directed, but not random migration, is cell cycle-dependent. In both 2D and 3D directed migration, cells in the G1 phase of the cell cycle outperformed cells in the G2 phase in terms of migration persistence and instantaneous velocity. These data suggest that in the presence of guidance cues in vivo, breast carcinoma cells in the G1 phase of the cell cycle may be more efficient in reaching vasculature.not peer-reviewed)