Characterizing Metastasis

Uncovering the drivers of metastasis using clinical data, RNA sequences, and Machine Learning.

Metastasis, the spread of cancer, is notoriously unpredictable: in some patients it emerges suddenly, in others it jumps to distant organs, Sometimes it remains dormant for years, sometimes it recurs after treatment, and in some rare cases it even regresses or disappears on its own. Yet, despite decades of research, we still lack a clear understanding of why or when each of these scenarios occurs.

We have assembled a large-scale dataset that integrates real-life stories of patients through their clinical data, and the molecular fingerprints of cancer through RNA sequencing of metastatic cells. By analyzing these side by side, we hope to uncover the hidden patterns that govern metastatic behavior. Specifically, we seek to identify the biological and clinical factors that predict whether metastasis will progress, lie dormant, recur, or resolve.

Our goal is simple but powerful: to transform metastasis from a mysterious, stochastic process into a predictable and potentially manageable one, opening pathways to personalized treatments and better prognostic tools that give patients better chances at fighting back.