Siqi Tan: ISEF Participant Explores Hidden Loops in Cancer- Persistent Homology Framework for PDAC Transcriptional State Discovery

Siqi Tan presenting at ISEF, Arizona, 2026

Pancreatic cancer remains deadly, with fewer than 1 in 8 patients surviving beyond five years. Current tumor cell analysis methods assume activity follows a straight or branching trajectory, potentially missing more complex patterns hidden in the data. Topological data analysis, specifically persistent homology, looks for loops in high-dimensional data, revealing patterns in gene activity that linear methods may overlook. Working at 46-fold finer resolution than standard clustering, the pipeline analyzed nearly 11,000 pancreatic cancer cells, uncovering a tumor-exclusive loop structure. Further analysis identified 135 drug-gene interactions with cyclic genes, several not previously associated with this type of recurring activity, suggesting new directions for how and when pancreatic cancer cells might be targeted.

Siqi Tan shares her thoughts about ISEF:

As for ISEF, it felt like a whole new world had opened up, with an unmatched excitement for science shared among the brightest young scientists in the world and opportunities to speak directly with industry professionals and renowned researchers. The experiences, lifelong friends, and mentors doing science fairs have allowed me are some of the most meaningful parts of the journey. I would not be the same person without science fair, not only in terms of skills but also in who I have become. I highly recommend it to any students aspiring to go into STEM fields!