A mechanistic understanding of cellular processes continues to be an elusive goal of
quantitative biology. Measurements across multiple spatiotemporal resolutions are
routinely collected but how these data impact our understanding of biological
processes is often not clear. In this presentation, I will address the link between
data and knowledge across multiple data types and explore the predictive power of
models given different types of data. I will first introduce various classifications of
data and how they relate to mechanistic explanations of cellular processes. We used
a Bayesian inference formalism to probe how different data types at different
resolutions can constrain model output and provide a probabilistic explanation for
network mechanisms. Finally, I will demonstrate this probabilistic approach to
understand network-driven processes in apoptosis signal execution, to identify
network execution modes and the impact of increased noise in these predictions.