Elucidating gene regulatory networks is crucial for understanding normal cell physiology and complex pathologic phenotypes. Existing computational methods for the genome-wide "reverse engineering" of such networks have been successful only for lower eukaryotes with simple genomes. Here we present ARACNE, a novel algorithm, using microarray expression profiles, specifically designed to scale up to the complexity of regulatory networks in mammalian cells, yet general enough to address a wider range of network deconvolution problems. This method uses an information theoretic approach to eliminate the majority of indirect interactions inferred by co-expression methods. We prove that ARACNE reconstructs the network exactly (asymptotically) if the effect of loops in the network topology is negligible, and we show that the algorithm works well in practice, even in the presence of numerous loops and complex topologies. We assess ARACNE's ability to reconstruct transcriptional regulatory networks using both a realistic synthetic dataset and a microarray dataset from human B cells. On synthetic datasets ARACNE achieves very low error rates and outperforms established methods, such as Relevance Networks and Bayesian Networks. Application to the deconvolution of genetic networks in human B cells demonstrates ARACNE's ability to infer validated transcriptional targets of the cMYC proto-oncogene. We also study the effects of misestimation of mutual information on network reconstruction, and show that algorithms based on mutual information ranking are more resilient to estimation errors. ARACNE shows promise in identifying direct transcriptional interactions in mammalian cellular networks, a problem that has challenged existing reverse engineering algorithms. This approach should enhance our ability to use microarray data to elucidate functional mechanisms that underlie cellular processes and to identify molecular targets of pharmacological compounds in mammalian cellular networks.

Cancer-associated fibroblasts (CAF) are major players in the progression and drug resistance of pancreatic ductal adenocarcinoma (PDAC). CAFs constitute a diverse cell population consisting of several recently described subtypes, although the extent of CAF heterogeneity has remained undefined. Here we use single-cell RNA sequencing to thoroughly characterize the neoplastic and tumor microenvironment content of human and mouse PDAC tumors. We corroborate the presence of myofibroblastic CAFs and inflammatory CAFs and define their unique gene signatures in vivo. Moreover, we describe a new population of CAFs that express MHC class II and CD74, but do not express classic costimulatory molecules. We term this cell population “antigen-presenting CAFs” and find that they activate CD4⁺ T cells in an antigen-specific fashion in a model system, confirming their putative immune-modulatory capacity. Our cross-species analysis paves the way for investigating distinct functions of CAF subtypes in PDAC immunity and progression. Significance: Appreciating the full spectrum of fibroblast heterogeneity in pancreatic ductal adenocarcinoma is crucial to developing therapies that specifically target tumor-promoting CAFs. This work identifies MHC class II–expressing CAFs with a capacity to present antigens to CD4⁺ T cells, and potentially to modulate the immune response in pancreatic tumors. See related commentary by Belle and DeNardo, p. 1001. This article is highlighted in the In This Issue feature, p. 983

The inference of transcriptional networks that regulate transitions into physiological or pathological cellular states remains a central challenge in systems biology. A mesenchymal phenotype is the hallmark of tumour aggressiveness in human malignant glioma, but the regulatory programs responsible for implementing the associated molecular signature are largely unknown. Here we show that reverse-engineering and an unbiased interrogation of a glioma-specific regulatory network reveal the transcriptional module that activates expression of mesenchymal genes in malignant glioma. Two transcription factors (C/EBPβ and STAT3) emerge as synergistic initiators and master regulators of mesenchymal transformation. Ectopic co-expression of C/EBPβ and STAT3 reprograms neural stem cells along the aberrant mesenchymal lineage, whereas elimination of the two factors in glioma cells leads to collapse of the mesenchymal signature and reduces tumour aggressiveness. In human glioma, expression of C/EBPβ and STAT3 correlates with mesenchymal differentiation and predicts poor clinical outcome. These results show that the activation of a small regulatory module is necessary and sufficient to initiate and maintain an aberrant phenotypic state in cancer cells.

Andrea Califano, Mariano Alvarez and colleagues present an approach, VIPER, for inferring protein activity in single cancer samples based on expression of a protein's downstream targets. The authors use VIPER to predict aberrant protein activity independent of mutational status and validate drug sensitivity predictions using in vitro assays. Identifying the multiple dysregulated oncoproteins that contribute to tumorigenesis in a given patient is crucial for developing personalized treatment plans. However, accurate inference of aberrant protein activity in biological samples is still challenging as genetic alterations are only partially predictive and direct measurements of protein activity are generally not feasible. To address this problem we introduce and experimentally validate a new algorithm, virtual inference of protein activity by enriched regulon analysis (VIPER), for accurate assessment of protein activity from gene expression data. We used VIPER to evaluate the functional relevance of genetic alterations in regulatory proteins across all samples in The Cancer Genome Atlas (TCGA). In addition to accurately infer aberrant protein activity induced by established mutations, we also identified a fraction of tumors with aberrant activity of druggable oncoproteins despite a lack of mutations, and vice versa. In vitro assays confirmed that VIPER-inferred protein activity outperformed mutational analysis in predicting sensitivity to targeted inhibitors.