Document Actions
List of Projects
ICBP Summer 2009 Research Program Project Descriptions
Dana-Farber Cancer Institute
Website: http://www.broad.mit.edu/cancer/collaborations/icbp/
Principal Investigator: Todd R. Golub, M.D.
Title: Functional Genomic Approaches to Cancer
Duration of Program: June 8 through August 7, 2009
The Dana-Farber/Broad Institute ICBP Center focuses on the use of emerging genomics methods to identify molecular ‘signatures’ in cancer cells. These signatures may be based on RNA or proteomic profiles, and have tremendous potential as new diagnostics and as a guide to the development of new therapeutic strategies for cancer. The ICBP summer training program will provide an opportunity for students to participate in the generation of the genomic data, and in the computational analysis of the results.
Duke University
Website: http://icbp.genome.duke.edu/
Principal Investigator: Joseph Nevins, Ph.D.
Co-Principal Investigator: Mike West, Ph.D.
Mentor: Lingchong You, Ph.D.
Title: Experimental Cancer Genomics and Statistical Modeling/Computational Analyses
Duration of Program: May 26 through July 31, 2009
The Duke Integrative Cancer Biology Program (ICBP) is focused on the development of experimental and computation methods to achieve a better understanding of the genes and pathways that result in cancer, specifically the gene regulatory mechanisms controlling cellular proliferation and the link with decisions of cell fate. These regulatory events are critical to normal development and the maintenance of tissue homeostasis and are the events often disrupted in the development of tumors. Our work primarily focuses on the retinoblastoma tumor suppressor (Rb) and the role of this protein in controlling the E2F transcription factor activities in the cell. Within this context, our work addresses the use of genome-scale measures of gene expression achieved with DNA microarray analysis to study the function of these signaling pathways. Additional work also makes use of gene expression profiling to develop more precise phenotypes of human cancers, including the prediction of clinical outcomes, and then link these more precise phenotypes with the action of oncogenic signaling pathways. Statistical and computational modeling research develops methods for exploratory and confirmatory analysis of genome-scale gene expression from both experimental studies and observational data from human cancers, aiming to aid in the investigation of substructure in these molecular pathways as well as their relevance to clinical outcomes.
Summer undergraduate research opportunities:
Students have opportunities to be involved in research experiences in experimental cancer genomics as well as statistical modeling/computational analysis of genomic data in pathway and outcomes studies.
E. O. Lawrence Berkeley National
Laboratory (LBNL)
Website: http://icbp.lbl.gov/index.htm
Principal Investigator: Joe Gray, Ph.D.
Duration of Program: June 8 through August 7, 2009
Project 1. Integration of OMIC with Cell Cycle Data
Mentor: Bahram Parvin, Ph.D.
The ICBP, at LBNL, has generated methylation, expression, and a number of biological endpoints. While array data (e.g., gene expression and methylation) represents base-line data for a panel of breast cancer cell lines, biological data are acquired through cell-based imaging and represented as cell cycle data. Currently, methylation and expression data have been integrated to infer a subset of genes that are epigenetically regulated. The student candidate will be tasked to develop computational methods to answer the following questions: (i) what methylation patterns , along the genome, are predictive of cell cycle data, (ii) which epigenetically regulated genes predict cell cycle data, and (iii) how many subpopulation of cell cycle data are there, and how do these subpopulations relate to basal and luminal lines.
Project 2. Identification of Predictive Biomarkers of Drug Response in Breast Cancer Cells and their Regulatory Controls
Mentor: Debo Das, Ph.D.
A rapidly growing number of drugs targeted against molecular defects in cancer are now being tested clinically. However, only a limited number of patients respond to such drugs. We are developing an in vitro systems approach to identify predictive markers that can stratify the tumor patients in terms of their drug response. The proposed summer project will focus on computational studies on identification of biomarkers and how they are regulated. The project will involve a combination of algorithm development, implementation and data analysis. Basic knowledge of statistics and programming skills are expected.
Massachusetts General Hospital/Harvard University
Websites: http://biosystems.mit.edu/
http://web.mit.edu/icbp/
http://web.mit.edu/dallab/
Principal Investigators: Thomas S. Deisboeck, M.D. (Mass Gen/Harvard)
Douglas Lauffenburger, Ph.D. (MIT)
Mentors: Thomas S. Deisboeck, M.D. (Mass Gen/Harvard)
Douglas Lauffenburger, Ph.D. (MIT)
Title: Integrative Brain Tumor Modeling
Location: Massachusetts General Hospital/Harvard University
and MIT
Duration of Program: June 8 through August 7, 2009
One undergraduate student will be invited to participate in collaborative biomedical research between the MGH ICBP, i.e. the Center for the Development of a Virtual Tumor (CViT; https://www.cvit.org) and the MIT ICBP. Led by Massachusetts General Hospital, CViT is a large-scale collaborative effort that focuses on building an international web-based community of cancer modelers including the development of a digital model repository. The purpose of the digital model repository is to provide investigators access to mathematical and computer models of cancer, as well as experimental and clinical data on cancer genes, proteins, tissues, and biological model systems. The MIT ICBP endeavor focuses on developing computational models in intimate association with quantitative experimental studies of regulatory networks governing cell death, proliferation, and migration behavior involved in tumor progression and treatment. Specifically, CViT collaborates with Massachusetts Institute of Technology's Department of Biological Engineering on applying a multiscale cancer modeling platform to experimental brain tumor cell data with a focus on determining how much single cell information is necessary to model multicellular cancer systems. While the experimental part is being conducted at MIT, the computational modeling part will be conducted at MGH. Mentored by both MGH CViT (Dr. Thomas S. Deisboeck) and MIT (Dr. Douglas Lauffenburger), this interdisciplinary training will include both "wet" laboratory sample handling (planning, and execution of in vitro experiments, and related image analysis methods) as well as computational analysis of the data.
Massachusetts Institute of
Technology
Websites: http://web.mit.edu/icbp/
http://web.mit.edu/dallab/
Principal Investigators: Douglas Lauffenburger, Ph.D. (MIT)
Thomas S. Deisboeck, M.D. (Mass Gen/Harvard)
Mentors: Douglas Lauffenburger, Ph.D. (MIT)
Thomas S. Deisboeck, M.D. (Mass Gen/Harvard)
Title: Integrative Brain Tumor Modeling
Location: Massachusetts General Hospital/Harvard University
and MIT
Duration of Program: June 8 through August 7, 2009
One undergraduate student will be invited to participate in collaborative biomedical research between the MGH ICBP, i.e. the Center for the Development of a Virtual Tumor (CViT; https://www.cvit.org) and the MIT ICBP. Led by Massachusetts General Hospital, CViT is a large-scale collaborative effort that focuses on building an international web-based community of cancer modelers including the development of a digital model repository. The purpose of the digital model repository is to provide investigators access to mathematical and computer models of cancer, as well as experimental and clinical data on cancer genes, proteins, tissues, and biological model systems. The MIT ICBP endeavor focuses on developing computational models in intimate association with quantitative experimental studies of regulatory networks governing cell death, proliferation, and migration behavior involved in tumor progression and treatment. Specifically, CViT collaborates with Massachusetts Institute of Technology's Department of Biological Engineering on applying a multiscale cancer modeling platform to experimental brain tumor cell data with a focus on determining how much single cell information is necessary to model multicellular cancer systems. While the experimental part is being conducted at MIT, the computational modeling part will be conducted at MGH. Mentored by both MGH CViT (Dr. Thomas S. Deisboeck) and MIT (Dr. Douglas Lauffenburger), this interdisciplinary training will include both "wet" laboratory sample handling (planning, and execution of in vitro experiments, and related image analysis methods) as well as computational analysis of the data.
The Ohio State University
Websites: http://icbp.med.ohio-state.edu/
http://mbi.osu.edu/
http://bioinformatics.med.ohio-state.edu/
Principal Investigator: Tim Huang, Ph.D.
Duration of Program: June 8 through August 7, 2009
Project 1
Mentor: Cenny Taslim, Ph.D.
Location: The Ohio State University
Title: Analysis of ChIP-Seq Data Using Finite Mixture Model
ChIP-Seq technology has enabled researchers to simultaneously study protein-DNA binding at whole genome scale with a higher accuracy comparing to ChIP-chip experiments. Analysis of ChIP-Seq data and identification of enriched regions of protein binding are important parts of ICBP research. The summer research fellow will assist in the identification of enriched regions by comparing various samples related to breast cancer. The candidate will learn how to perform data normalization using loess smoother and identify the enriched regions using finite mixture model. Normalization is a very important step in analyzing ChIP-Seq data due to various biases and noises such as imaging, sequencing, mapping and background noise. Normalization helps reduce the effects of these noises. After normalization, finite mixture modeling approach is applied to find a model which represents the data best. EM (Expectation Maximization) algorithm is used to estimate the parameters of the finite mixture model. Classification based on the best model found using finite mixture modeling approach is implemented to identify the enriched regions. The candidate will help in performing experiments to optimize the estimation process and conduct performance benchmarking with existing ChIP-Seq analysis methods.
Project 2
Mentors: Kenneth P. Nephew, Ph.D. and Sun Kim, Ph.D.
Location: Indiana University
Title: Building Epigenetic Models of Drug Resistant Cancer
Tumors contain a small population of “cancer stem/initiating cells” that propagate the entire tumor. While conventional cancer therapies are effective at killing the majority of cells within the tumor, they may miss these cancer stem cells. Epigenetic changes (changes in DNA structure) may be a way to characterize cancer stem cells and build models of epigenetic changes that contribute to drug resistant disease. Furthermore, altering those DNA structural changes may represent a new type of therapy for targeting and killing cancer stem cells. Directly targeting such tumor precursor cells could halt tumor growth and perhaps completely eliminate the tumor.
Recently, we have isolated and characterized putative ovarian cancer stem/initiating cells from patient tumors. We believe it is highly likely that epigenetic changes can affect multiple pathways in these cancer initiating cells, termed OCICs (ovarian cancer initiating cells), including pathways that contribute to drug resistance. During the training period, the summer ICBP Fellow will test the hypothesis that epigenetic events in ovarian cancer stem cells contribute to response to chemotherapy and ovarian cancer drug resistance. To reveal “epigenetic signatures” underlying the molecular complexity of chemoresistance, the ICBP Summer Fellow will assist in conducting a comprehensive pattern analysis of ovarian cancer stem/initiating cells using massive parallel bisulphite pyrosequencing (454 LifeSciences and described on our website: http://cgb.indiana.edu/genomics/genome_sequencing/). The mentored project will generate a consensus sequence of the ovarian cancer stem/initiating cells that is derived from the assembly (de novo sequencing) or mapping (resequencing) software of the 454 instrument. Predictive models for methylation susceptibility will be investigated utilizing CpG site specific methylation levels; flanking sequences will be used to characterize methylation susceptibility in terms of character compositions and build predictive models for DNA methylation susceptibility. In addition, the project will include characterization of ovarian cancer stem/initiating cells methylation pattern signatures and related sequence and machine learning analysis. In summary, we believe that these computational approaches will allow the generation of “epigenetic models” of ovarian cancer drug resistance that could provide for novel approaches for managing this devastating clinical phenomenon. Reversal of such events may allow for the “epigenetic resensitization” of resistant cells/tumors.
Stanford University
School of Medicine
Website: http://icbp.stanford.edu
Principal Investigator: Sylvia Plevritis, Ph.D.
Mentor: Dean Felsher, M.D.
Title: Mouse Models of Cancer: Oncogenic Inititation of Tumorigenesis
Duration of Program: June 8 through August 7, 2009
This summer project will expose the student to mouse models of cancer, and will be hosted by the Felsher laboratory at Stanford University.
(http://med.stanford.edu/labs/dean_felsher/):
The focus is on how oncogenes such as MYC initiate and maintain tumorigenesis. We have developed model systems whereby we can conditionally activate oncogenes in normal human and mouse cells in tissue culture or in specific tissues of transgenic mice.
In particular, using the tetracycline regulatory system, we have generated a conditional model system for MYC-induced hematopoietic tumors. Using the tet system, we have shown that cancers caused by the conditional over-expression of the MYC proto-oncogene regress with its inactivation. Thus, even though cancer is a multi-step process, the inactivation of one oncogene can be sufficient to induce tumor regression. Now, we are using these model systems to address three
questions:
1. How do oncogenes initiate tumorigenesis?
2. How does oncogene inactivation cause tumor regression?
3. How do tumors escape dependence on oncogenes?
Applicants should have lab experience and have taken courses in molecular biology. There will also be opportunity to participate in computational analysis of high-throughput data.
University Hospital Case Medical Center
Website: http://www.case.edu/med/icbp/
Principal Investigator: Tim Kinsella, M.D.
Mentors: Ken Loparo, Ph.D. and Tim Kinsella, M.D
Duration of Program: June 8 through August 7, 2009
Project 1: Modeling of DNA Mismatch Repair Pathway
Location: University Hospital Case Medical Center
The DNA mismatch repair (MMR) pathway is an important repair mechanism in the cell that ensures genomic stability. Mismatch repair deficiencies are shown to be associated with certain hereditary forms of cancer as well as many sporadic cancers. The loss of mismatch repair also leads to resistance to chemotherapeutic agents and other types of DNA stress including ionizing radiation.
The aim of this research project is to develop a stochastic hybrid model to study the dynamics of the mismatch repair pathway and for the development of therapeutic strategies. The experimental data required to support the hybrid modeling effort are currently being collected from in vitro wet lab experiments. The analysis of the hybrid model both analytically, and by simulation (in silico experimentation) will provide insights into the dynamics of the mismatch repair pathway. Understanding the dynamics will provide a foundation for more detailed control studies where the objective is to synthesize dosing strategies of chemotherapeutic agents and radiation that increases therapeutic gain for individual patients.
Project 2: Modeling of Base Excision Repair Pathway
Location: University Hospital Case Medical Center
Base excision repair (BER) is a repair pathway that can contribute to the effectiveness of cancer treatment through chemotherapy. Alkylating agents are a major class of therapeutic drugs and one function of the base excision repair pathway is to reduce the cytotoxicity of these agents by correcting the induced damage. The relevant literature shows that inhibition of the BER pathway can result in increased sensitivity to these alkylating agents, thereby making it an attractive target for anti-cancer therapy. Methoxyamine is one of the inhibitors that is currently being studied.
The aim of this research project is to develop a stochastic hybrid model of the base excision repair process with inhibition in order to study the underlying dynamics of the BER process. The efficiency of the inhibition plays an important role in improving the therapeutic efficacy of anticancer agents. When drug and inhibitor combinations are used, it is crucial to determine the optimal doses for both in different cancer cell lines, and ultimately to determine the patient specific timing and dosing of the agents that will have the best therapeutic effect. A comprehensive dynamical model of the BER process will enable and facilitate the study of the time course of the BER process and the influence of inhibition on this process. The analysis and simulation of these models will enhance the understanding of how the process and inhibition work collectively in different cancer cell lines, and how the process can be manipulated using inhibitors to maximize the efficacy of the chemotherapeutic agents.
Project 3: Laboratory Investigations Linking DNA Mismatch Repair Processing of Chemotherapy Damage to Cell Death Pathways
Location: Stony Brook University
DNA mismatch repair (MMR) is now recognized to be involved in processing DNA damages by several classes of active chemotherapy drugs. My laboratory has shown the MMR processing of 6-thioguanine (6-TG) results in a prolonged G2 cell cycle delay activated by ATR-Chk1 pathway activation. Subsequently, both type 1 programmed cell death (apoptosis) and type 2 programmed cell death (autophagy) pathways are activated.
The goal of this summer research project will be to participate in ongoing molecular, biochemical and cellular studies of 6-TG processing by MMR in isogenic human tumor cell lines to improve our understanding of these two cell death pathways. Wet lab exposure to basic techniques including tissue culture, flow cytometry, western blotting and immuno-fluorescence will be available to the summer fellows. Additionally, exposure to mathematical and/or computational modeling of these data will be provided.
Vanderbilt University Medical Center
Website: http://www.vanderbilt.edu/VICBC/
Principal Investigator: Vito Quaranta, M.D.
Mentors: Darren Tyson, Ph.D. and Mohamed Hassanein, Ph.D.
Title: Study of Cancer Growth and Invasion: Data Acquisition of Single Cell Core Processes and Implementation into Mathematical Models
Duration of Program: June 1 through Friday July 31, 2009
Our emphasis is in studying cancer using tools of mathematical and computational modeling. As cancer biologists, we experiment both in cell lines (normal and cancer) and animal models to extract the important information required in the mathematical models that we are building with our collaborators at the Moffitt Cancer Center, Tampa, FL. The particular techniques that the candidates would expect to learn include: cell culture of normal and cancer cells, in vitro assays of cell growth, cell death, cell migration, cell-matrix interactions, cell-cell adhesion, cellular metabolism, all important parameters in cancer. These methods will focus on single-cell measurements of these processes and include live cell imaging, fluorescent functional assays, flow cytometry, among others. In addition, the student would have the opportunity to acquire experience in the emerging field of Integrative Cancer Biology.
