FAQ for Rule-based Modeling of Cell-level Networks in Immuno-oncology

Abstract

Identifying how proteins secreted by malignant cells reorganize the composition and functional orientation of different cells with a tissue during oncogenesis has important implications for developing therapies for cancer. Given the inherent dynamics and complexity associated with the interplay between secreted proteins and networks of cell-to-cell communication within tissues, mechanistic mathematical modeling and simulation is becoming an important tool in discovering and developing anti-cancer therapies. Currently, the topology or structure of the model, which includes the cell types (i.e., nodes) and their interactions (i.e., causal arcs), is created by hand. As hand-curated models can implicitly impose bias on how data is interpreted, we have prototyped a data-driven, rule-based modeling approach for identifying these cell-level networks in the context of immuno-oncology. While the initial effort focused on predicting the cell-level network based on human data and validating aspects of the predicted network using syngeneic mouse models, we can envision a project that can test this approach more rigorously. However, there are a number of questions that people have raised about this approach. Here, I'm going to address some of these frequently asked questions (FAQs).