A key to understanding complex biological systems such as cancer is to understand how networks of molecular interactions function in normal cells, and how they becomeperturbed in disease states. This endeavor demands sophisticated computational tools to derive and analyze molecular regulatory networks. Such tools need to integrate multiple types of data from high-throughput experiments to derive specific predictions - a systems biology approach. In this proposal we describe the development of novel machine learning methods to infer regulatory networks, that integrate known biology (prior information) with multiple types of high-throughput data. Our methods for inferring regulatory networks will be a key component of each biological project, and there will be an iterative feedback between the experimental and computational components to refine models and suggest new rounds of experiments. We will also develop computationally executable biological models that allow us to pose hypothetical "what if" questions to predict how a system will respond if we perturb it in a specific way, such as a therapeutic intervention. Phenotypic responses in cells can often be attained via more than one route, which is particularly problematic in cancer. A therapeutic regime may target a particular weakness in tumor cells, only for them to acquire additional mutations that elude the drug intervention by employing alternative pathways to achieve the same outcomes of survival and proliferation. We will develop multiscale models of cancer that estimate cellular level dynamics from tumor growth kinetics. Systems biology has benefited from utilizing methods originally developed in other disciplines for very different reasons. Major contributions have been made by applying methodologies from mathematics, statistics, computer science, engineering, and physics to biological questions. We plan to extend this interdisciplinary reach by adopting and developing ideas from the mathematics of geometry and topology in the identification of underlying patterns of progression in cancer, and to the analysis of large molecular networks.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Specialized Center--Cooperative Agreements (U54)
Project #
Application #
Study Section
Special Emphasis Panel (ZCA1-SRLB-C)
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Stanford University
United States
Zip Code
Gallaher, Jill; Babu, Aravind; Plevritis, Sylvia et al. (2014) Bridging population and tissue scale tumor dynamics: a new paradigm for understanding differences in tumor growth and metastatic disease. Cancer Res 74:426-35
Bendall, Sean C; Davis, Kara L; Amir, El-Ad David et al. (2014) Single-cell trajectory detection uncovers progression and regulatory coordination in human B cell development. Cell 157:714-25
Sinha, Subarna; Tsang, Emily K; Zeng, Haoyang et al. (2014) Mining TCGA data using Boolean implications. PLoS One 9:e102119
Choi, Peter S; Li, Yulin; Felsher, Dean W (2014) Addiction to multiple oncogenes can be exploited to prevent the emergence of therapeutic resistance. Proc Natl Acad Sci U S A 111:E3316-24
Gaudillière, Brice; Fragiadakis, Gabriela K; Bruggner, Robert V et al. (2014) Clinical recovery from surgery correlates with single-cell immune signatures. Sci Transl Med 6:255ra131
Sen, Nandini; Mukherjee, Gourab; Sen, Adrish et al. (2014) Single-cell mass cytometry analysis of human tonsil T cell remodeling by varicella zoster virus. Cell Rep 8:633-45
Li, Y; Casey, S C; Felsher, D W (2014) Inactivation of MYC reverses tumorigenesis. J Intern Med 276:52-60
Li, Yulin; Choi, Peter S; Casey, Stephanie C et al. (2014) Activation of Cre recombinase alone can induce complete tumor regression. PLoS One 9:e107589
Bruggner, Robert V; Bodenmiller, Bernd; Dill, David L et al. (2014) Automated identification of stratifying signatures in cellular subpopulations. Proc Natl Acad Sci U S A 111:E2770-7
Gevaert, Olivier; Mitchell, Lex A; Achrol, Achal S et al. (2014) Glioblastoma multiforme: exploratory radiogenomic analysis by using quantitative image features. Radiology 273:168-74

Showing the most recent 10 out of 31 publications