Abstract Pancreatic cancer is characterized by an expansive, desmoplastic stroma that responds to neoplastic signals from the malignant epithelium while also instigating a cacophony of signals from diverse stromal cell types that in turn modulate the malignant process. Several published efforts have reported the global signaling profiles of bulk pancreatic tumors, which have proven useful for defining different molecular subtypes of the disease. However, in order to facilitate the precise assessment of both neoplastic and stromal signaling profiles, we developed a robust pipeline to perform RNA-sequencing from laser-capture microdissected tissues. We applied this technique to a collection of 200+ frozen primary pancreatic ductal adenocarcinomas, deriving both malignant epithelial and stromal gene expression profiles from each. These compartment-specific profiles may serve to define novel molecular subtypes grounded in the biology of the distinct compartments. This technique also enabled profiling of the epithelium and stromal from microscopic PanIN-1 proliferations, facilitating the precise delineation of changes arising during malignant progression. Using computational modelling of paired epithelial and stromal samples, we developed a novel digital deconvolution model that generates virtual compartment-specific gene expression signatures from bulk pancreatic tumor profiles. Thus, findings from our experimentally dissected sample set can be leveraged to analyze multiple publically-available gene expression datasets. In order to learn more about the regulatory processes that drive malignancy in both the epithelial and stromal compartments, we used a systems biology approach to reconstruct their global regulatory networks in pancreatic cancer. Briefly, regulatory models comprising ~700,000 interactions were computationally reconstructed for each compartment from the LCM-RNA-seq datasets using the ARACNe algorithm. These networks then served as a scaffold for the application of the MARINa and VIPER algorithms, which define master regulatory proteins whose activity drives (or restrains) the PanIN-1 to PDA progression. We will discuss early results of master regulator analyses that illuminate both oncogenic signaling in the epithelium, and large-scale changes in stromal microenvironment that occur during progression from PanIN-1 to PDA. Citation Format: H. Carlo Maurer, Jing He, Mariano J. Alvarez, Federico Giorgi, Sam Holmstrom, Andrea Califano, Kenneth P. Olive.{Authors}. Reconstructing compartment−specific regulatory programs in pancreatic cancer. [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer: Advances in Science and Clinical Care; 2016 May 12-15; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2016;76(24 Suppl):Abstract nr IA16.