The efforts of our MIT ICBP, entitled the 'Tumor Cell Networks Center' (TCNC), focus primarily on cancer biology, with intersection into experimental therapeutics. The TCNC comprises three multi-investigator research projects and cores in Computational Modeling and in Education/Outreach.
Each of the three research Projects involves a combination of genomic- and/or proteomic-based experimental studies with computational analysis and modeling. The Projects aim to provide novel biological insights and make testable predictions concerning the modulation and dysregulation of molecular networks in cancer biology and treatment. The Modeling Core is focused on developing improved computational methods for use not only by TCNC investigators but also by the broader cancer community.
A central thread integrating our experiment and modeling efforts is a paradigmatic 'cue-signal-response' framework for probing and modeling the regulation of tumor cell phenotypes critical to tumor cell biology. We seek to understand the convolution of genotype and environment in controlling molecular networks and consequent phenotypes. Environmental factors are cast in the 'cue' category, comprising biochemical ligands (cytokines, extracellular matrix, inhibitory drugs) as well as physical inputs such as DNA damaging agents (radiation, chemical toxins). Intracellular regulatory states generated by exposure to these cues are cast in the 'signal' category; in our work these primarily involve protein signaling and gene expression pathways. Resulting phenotypic behaviors, such as cell death, proliferation, and migration, are cast in the 'response' category, as executed by cytoskeletal, metabolic, and transcription/translation processes. To achieve the CCSB mandate of "integration of experimental and computational approaches towards the understanding of cancer biology", we aim to develop and make available to the cancer biology community successful models of three distinct but inter-related classes:
A. 'Cell Response' models, capable of predicting how cellular phenotypes are executed by biophysical molecular processes;
B. 'Signal-Response' models, capable of predicting how cellular responses are regulated by information carried by intracellular signaling pathways; and
C. 'Cue-Signal' models, capable of predicting what intracellular signals are generated by environmental stimuli and therapeutic agents in the context of specific genotypes.
The construction and testing of all three classes of models is currently underway and linked to three important problems in cancer biology and therapy: (a) cancer progression via dysregulation of mitogenic signaling pathways downstream of receptor tyrosine kinases (with particular focus on the ErbB receptor system); (b) cancer progression via inappropriate cell migration processes that promote invasive and metastatic behavior; and (c) cancer treatment via molecular pharmaceuticals including chemotherapeutics and targeted inhibitors (small molecules and antibodies). We believe that there is no "one size fits all" modeling approach suited to analyzing, interpreting, and/or predicting all types of experimental observations or answering all types of biological questions in the realm of these problems - just as there is no "one size fits all" experimental assay or context suited to generating all kinds of experimental data. Thus, our experimental studies include cell culture and mouse contexts, along with human tissue samples, and our modeling approaches span a wide spectrum of methods aimed at identifying correlative relationships (e.g., partial least-squares regression), network topologies (e.g., Steiner Forest algorithm), influences of one component on another (e.g., Boolean and fuzzy logic, Bayesian networks), and physicochemical mechanisms (differential equations).
The goal of the Mitogenic Networks Project is the development of high-level statistical and specific physico-chemical models that describe key features of mitogenic signaling networks activated by ErbB receptors and by oncogenic K-ras.
The Migration Networks Project is designed to understand how key steps in the onset and progression of metastatic disease are regulated. Our overall goal is to identify new targets for therapeutic intervention in metastatic disease, the most frequent cause of mortality in cancer patients.
DNA Damage Networks
The goal of the DNA Damage Networks project is the development of statistical and physico-chemical models that describe the cellular response to DNA damage at the network and systems biology level.