Inferring Dynamic Cellular Trajectories and Underlying Cellular Regulatory Networks with Neural and Graph ODE Models - Part 2

Learn advanced computational methods for analyzing cellular dynamics and regulatory networks in this second part of a two-part tutorial presented at IPAM's Mathematics of Cancer workshop. Explore the RITINI (Regulatory Interaction Network Inference) graph ODE framework for inferring regulatory mechanisms underlying dynamic cellular trajectories, building upon the PHATE embedding method and MIOflow neural ODE framework covered in part one. Examine a detailed case study in breast cancer research that demonstrates how cells transform from cancer stem cell states to aggressive metastatic states, and discover how these computational approaches can identify potential state-gating targets for therapeutic intervention. Gain insights into cutting-edge mathematical modeling techniques that combine data diffusion geometry, neural ODEs, and graph-based methods to understand complex biological processes at the cellular level.