In this work, we collaborated with prof. Ilan Tsarfaty’s lab in order to leverage Deep Learning techniques for the labs’ research efforts. Segmenting and tracking (S&T) single cell nucleus in high resolution videos, and time-series clustering in order to identify the different behaviors of metastatic cells with different treatments. This kind of collaboration has altered our focus from deep dive into the newest methods in the DL domain of research and sent us to explore and use tools that are utilizing DL in order to improve some of the research done in the lab on real data. After extensive research and collaboration with Ilans’ lab we were able to improve the lab’s S&T pipeline in a significant order - tracking cells for much longer periods, providing new possibilities for future experiments. As we ran an experiment from the lab through our new pipeline, we set out to discover different patterns of appearance and motion (morpho-kinetic) features using techniques from the deep-learning toolkit. This approach, using only the data to guide us, wasn’t successful in providing new insights. Nevertheless, the approach for applying deep learning on similar domains proved to be insightful, and we are hoping that further collaboration with the lab will improve our results.