Seminars/Events
COMPLETE 2007/2008 SEMINAR SCHEDULE (revised April 17th, 2008)UPCOMING SEMINARS - SPRING 2008:
Student Research Seminars - May 30th, 2008
2:00pm – COOK 3118 A&B
Structural Evaluation of Bound, Unbound and Mutational Versions of Epstein-Barr Virus glycoprotein 42
Pamela Shaw, M.S., CBB Master's Student
ABSTRACT: With the recent solution of the unbound structure for EBV gp42, new opportunities have arisen to evaluate the changes this protein adopts between its unbound and bound-to-HLA Class II states. It also sheds light on the reasons that some mutational forms of gp42 are expressed at the cell surface but do not functionally operate in fusion. This current study evaluates the differences between bound and unbound gp42, and analyzes some structural possibilites in selected mutational states.
Research Advisor:
Richard M. Longnecker, Department of Microbiology-Immunology, Feinberg School of Medicine
The modularity landscape of networks
Kenji Wellman, CBB Master's Student
ABSTRACT: Modeling networks is of fundamental importance to computational systems biology. Biological examples include metabolic networks, the signal transduction networks, and neurological networks (Newman & Girvan, 2004). Such networks can include ones inferred from large amounts of high-throughput data, or pieced together from many years of experimental data (for instance the KEGG database). Many of these networks divide naturally into modules in which there is a high degree of connectivity between the nodes, also known as the "community structure" (Guimera & Amaral, 1995). These clusters could represent functional units of the system (Guimera & Amaral, 1995). Extracting this structure could provide useful information that would aid in inferring evolutionary relationships and the function of the components in the system, or in predicting behavior of the system under different environmental cues (Newman & Girvan, 2004). For instance, the modularity of a system is an important characteristic that affects the spread of disease in a network (Watts and Strogatz, 1998). Additionally, possible drug targets can be evaluated in the context of the module in which they belong.Research Advisor:
Research advisor:
Luis A.N. Amaral, Chemical & Biological Engineering, McCormick School of Engineering
Student Research Seminars - June 6th, 2008
2:00pm – COOK 3118 A&B
Structure-Activity Relationship Study of MEK 4 with Genistein
Sankar Narayan Krishna, CBB Master's Student
ABSTRACT: Dual specificity mitogen-activated protein kinase kinase 4(MEK4) regulates prostate cancer (PCa) cell invasion and metastasis. Inhibition of MEK4, by Genistein, occurs by blocking the P38 Map Kinase pathway in turn leading to inhibition of metastasis in prostate cancer cells. In my study, the goal is to understand the interactions of Genistein with the MEK-4 active site and to be able to use Structure-Activity relationship information derived from the study for the synthesis of a more effective lead compound. I previously created homology models of MEK4 protein and will use these models to propose "in-silico" docking interactions with Genistein and select modified analogues of the same. The resulting models will be used to study binding site interactions and to predict analogues of Genistein with higher efficacy than the parent molecule.
Research Advisors:
Raymond Bergan, Department of Medicine, Feinberg School of Medicine
Wayne Anderson, Molecular Pharmacology and Biological Chemistry, Feinberg School of Medicine
Motif Discovery in DNase Hypersensitive Sites
Jairav Desai, CBB Master's Student
ABSTRACT: Cystic fibrosis (CF) is a hereditary disease of complex pathology that primarily affects the lungs and digestive system resulting in a progressive, life shortening disability. The cystic fibrosis transmembrane conductance regulator gene (CFTR), when mutated, causes cystic fibrosis and has thus has been highly investigated. The gene shows a tightly regulated pattern of expression; the molecular basis for which is not understood. It is clear that important cis regulatory elements are not just restricted to the gene's promoter region. The Harris lab has identified regulatory elements relevant to the CFTR gene and mapped the function of these elements using comparative genomics; in particular, data obtained from the ENCODE (ENcyclopedia of DNA Elements) database was used to identify potential DNA regulatory regions in CFTR using data from multiple experiments in a number of different cell lines. One particularly powerful technique, the DNAse CHIP, identifies DNAse Hypersensitive Sites (DHS) in DNA which indicate nucleosome-depleted regions of chromatin and thus potential sites for the binding of regulatory elements. The results of DNase CHIP analysis on regions of the CFTR gene identified from the ENCODE database will be used to identify motifs for potential transcription factor binding sites and de novo motifs within DHS. Additional cell types have been analyzed which will be discussed in this presentation as well. Another major area of interest is our ongoing attempt to develop a CFTR-centric GBrowse to act as a central repository of information for the Harris Lab.
Research Advisor:
Ann Harris, Department of Pediatrics, Children´s Memorial Research Center
PAST SEMINARS:
Synchronized firing and visual signaling in the primate retina
E.J. Chichilnisky, PhD; Associate Professor
Systems Neurobiology Laboratories, Salk InstituteFriday, May 23rd, 2008 – 2:00pm
COOK 3118 A&B
Videocast in the Ward Building, room 5-230, on the Chicago campus
ABSTRACT: Synchronized firing has been hypothesized to play a role in the function of many neural circuits. In the retina, synchronized firing has been observed in several species, but its impact on visual signals remains unclear. First, no previous studies have examined synchronized firing in the retina of primates, the dominant model for human visual system function. Second, synchronized firing has been studied almost entirely on the basis of recordings from pairs of cells, but may occur at a much larger scale. Third, capturing the full structure of synchronized firing and its impact on visual processing presents substantial analytical challenges. As a consequence, the overall structure of synchronized firing and its impact on visual signals remains unclear. Using large-scale recordings from the primate retina in vitro, we have characterized the full structure of synchronized firing in populations of hundreds of retinal ganglion cells. Using two novel analysis methods, we find that the large-scale structure of synchronized firing can be explained by simple interactions between neighboring cells in the network. Further, accounting for this structure in decoding the retinal signal can yield up to 20% improvement in visual performance. Thus, synchronized firing in the primate retina may be understood with simple models and may represent a significant aspect of visual signaling to the brain.
Student Research Seminars - May 16th, 2008
2:00pm – COOK 3118 A&B
Direct-to-Consumer Personal Genome Service: A review of its nature, availability and public perception
Chia-Ju Lee, CBB Master's Student
ABSTRACT: Direct-to-Consumer (DTC) personal genome service is a burgeoning business where a novel type of genetic testing service is provided. The business commercializes the advanced technologies of genome-wide profiling of genetic variants used by researchers to find association of these variants with common, yet complex diseases, health-related conditions, or non-health related traits. Furthermore, the businesses advertise and sell services directly to the general public; the consumer orders the test directly and receives the results without the involvement of an independent health care provider serving as an intermediary. Following the first official launch of the service on November 6, 2007, a handful of startups rushed into the market. The highly science-and-technology-based industry appears to be a controversial entity. Pros and cons are prevalent in the press, blogs and the academic community. This project aims to systematically examine the nature of DTC personal genome service in terms of the science and technology behind the service, potential benefits and disadvantages, appropriateness of the service evaluated by the relevant professional societies, and governmental regulation and oversight with respect to consumer protection. In addition, a preliminary review of the availability of and the public’s perception of the service is conducted to explore future research topics on the ethical, economic, legal or social impact entailed by the emergence of the business.
Research Advisors:
Simon Lin, Basic Sciences,Feinberg School of Medicine Hongmei Jiang, Statistics, Weinberg College of Arts and Sciences
Independent Component Analysis of fMRI Acupuncture with attention to Auditory and Visual Cortex
Omar Akbik, CBB Master's StudentABSTRACT: After introducing the basic mechanics of fMRI and reviewing the methodology of acupuncture, Independent Component Analysis (ICA) will be reviewed and applied to the data set. The strengths and weakness of such an analysis for this particular type of fMRI treatment will be reviewed as well non-acupuncture treatment. The discussion will conclude with review of the results, areas of activation, as well as suggestions for the next stage of experiments.
Research Advisors:
Todd Parrish, Department of Radiology, Feinberg School of Medicine
JiPing Wang, Department of Statistics, Weinberg College of Arts and Sciences
Simultaneous estimation of trees and multiple sequence alignments
Tandy Warnow, PhD; Professor
Department of Computer Science, University of Texas &ndash AustinFriday, May 2, 2008 – 2:00pm
COOK 3118 A&B
ABSTRACT: The inference of evolutionary history presents enormous computational and mathematical challenges which are particularly exacerbated when dealing with very large datasets (containing thousands of sequences) or when sequences evolve under complex models of evolution. In this talk, I will describe some of the recent progress in large-scale phylogenetics, focusing on simultaneous estimation of multiple sequence alignments and phylogenies.
Student Research Seminars - April 11th, 2008
2:00pm – COOK 3118 A&B
Structure-Activity Relationship Study of MEK-4 & Genistein
Sankar Narayan Krishna, CBB Master's Student
ABSTRACT: Dual specificity Mitogen-activated protein kinase kinase 4(MEK-4) regulates Prostate Cancer (Pca) cell invasion and metastasis. Inhibition of MEK-4 by the indigenously developed lead compound from the Bergan Lab, Genistein, occurs by blocking the P38 MAP kinase pathway, in turn leading to inhibition of metastasis in prostate cancer cells. In my study, the goal is to understand the interactions of Genistein with the MEK-4 active site and to use the derived structure-activity relationship information for the synthesis of a more effective lead compound. The study will initially use computational approaches, homology modeling of the MEK-4 protein and docking simulation of MEK-4 with Genistein. Additionally, the modeling will be subject to intense scrutiny using validated experimental data and only then be used for prediction purposes. The predicted agents will then be tested and used during the drug synthesis stage.
Research Advisors:
Raymond Bergan, Department of Medicine, Feinberg School of Medicine
Wayne Anderson, Molecular Pharmacology and Biological Chemistry, Feinberg School of Medicine
Characteristics of Epstein-Barr Virus Envelope Protein gp42: Application of Sequence and Structure Analysis
Pamela Shaw, M.S., CBB Master's Student
ABSTRACT: Epstein-Barr virus (EBV) is a gamma herpes virus which requires a number of envelope glycopyroteins for fusion and entry into host cells. Human EBV gp42 is required for infection of host B-lymphocytes, but it is inhibitory to entry into epithelial cells. EBV gp42's crystalline structure, as bound to its HLA Class II receptor, was determined in 2002. The structure reveals that gp42 contains a canonical natural-killer receptor-like C-type lectin-like domain (NK-receptor-like CTLD). Using features of gp42, we have began our analysis of gp42 by comparing it to other species viral proteins; investigating its canonical CTLD domains and attempting to predict the structure of the flexible N-terminal region implicated in binding to other EBV envelope glycoproteins. Such an understanding of protein structure and sequence assists future mutational studies which will ultimately lead to improved methods of inhibiting EBV fusion and entry to host cells.
Research Advisor:
Richard M. Longnecker, Department of Microbiology-Immunology, Feinberg School of Medicine
Translational Systems Biology: Connectivity, Network Models and Panel Biomarker Discoveries
Jake Yue Chen, PhD; Assistant Professor
Department of Informatics & Computer and Information Science Indiana University – Purdue University IndianapolisABSTRACT: Synchronized firing has been hypothesized to play a role in the function of many neural circuits. In the retina, synchronized firing has been observed in several species, but its impact on visual signals remains unclear. First, no previous studies have examined synchronized firing in the retina of primates, the dominant model for human visual system function. Second, synchronized firing has been studied almost entirely on the basis of recordings from pairs of cells, but may occur at a much larger scale. Third, capturing the full structure of synchronized firing and its impact on visual processing presents substantial analytical challenges. As a consequence, the overall structure of synchronized firing and its impact on visual signals remains unclear. Using large-scale recordings from the primate retina in vitro, we have characterized the full structure of synchronized firing in populations of hundreds of retinal ganglion cells. Using two novel analysis methods, we find that the large-scale structure of synchronized firing can be explained by simple interactions between neighboring cells in the network. Further, accounting for this structure in decoding the retinal signal can yield up to 20% improvement in visual performance. Thus, synchronized firing in the primate retina may be understood with simple models and may represent a significant aspect of visual signaling to the brain.
Student Research Seminars - March 7th, 2008
2:00pm – COOK 3118 A&B
Motif Discovery in DNase Hypersensitive Sites
Jairav Desai, CBB Master's Student
ABSTRACT: Cystic fibrosis (CF) is a hereditary disease of complex pathology that affects mainly the lungs and digestive system and results in progressive, life shortening disability. The cystic fibrosis transmembrane conductance regulator gene (CFTR), when mutated, causes cystic fibrosis and is thus has been highly investigated. The gene shows a carefully regulated pattern of expression; the molecular basis for which is not understood. It is clear, however, that important cis regulatory elements are not just restricted to the gene´s promoter region. The Harris lab has identified regulatory elements relevant to the CFTR gene and mapped the function of these elements using comparative genomics; in particular data obtained from the ENCODE (ENcyclopedia of DNA Elements) database was used to identify potential DNA regulatory regions in CFTR using data from multiple experiments in a number of different cell lines. One particularly powerful line of investigation, DNAse CHIP, identifies DNAse Hypersensitive Sites (DHS) in DNA, which indicate nucleosome-depleted regions of chromatin and thus potential sites for the binding of regulatory elements. DHS CHIP data using DNA regions of the CFTR gene identified from the ENCODE database will be used to elucidate motifs for potential transcription factor binding sites and de novo motifs within DHS.
Research Advisor:
Ann Harris, Department of Pediatrics, Feinberg School of Medicine
An Investigation of the Oncomine User Interface as a Model for Creating New Microarray Data Mining Tools
Chia-Ju Lee, CBB Master's StudentABSTRACT: Extremely large amounts of microarray data have been produced and accumulated. The databases of the two largest repositories for microarray data have grown exponentially each year. As of October 2007, 173,486 arrays have been deposited into the NCBI GEO database and 97,635 arrays deposited into the ArrayExpress (EMBL-EBI). This data will gain additional value using meta-analysis, combining the results of several different studies which address related research hypotheses. A user-friendly web-based interface is an essential part of a data mining tool to help users effectively explore thousands of experiments and tens of millions of gene expression profiles stored in databases. The study will investigate the user interface of Oncomine, in terms of the principles that may be considered during the design of a more user-friendly interface.
Research Advisor:
Simon Lin, Basic Sciences, Feinberg School of Medicine
Scanning the Human Genome at Kilobase Resolution
San Ming Wang, PhD; Assistant Professor
Department of Medical Genetics, Feinberg School of MedicineFriday, February 22nd, 2008 – 2:00pm
COOK 3118 A&B
ABSTRACT: Increasing evidence shows that the genome structure is highly variable within the normal human population, and the genome structure in pathological situations, such as in cancer, is also highly altered, reflecting the heterogeneous and progressive nature of disease cells. Systematic analysis of normal genome variation will provide fundamental knowledge to improve understanding of the genetic basis for normal human diversity; comprehensive characterization of pathological genome alterations will identify genetic factors which contribute to particular diseases, which could become diagnostic markers and therapeutic targets. Structural genome changes frequently happen at kilobase to sub-kilobase levels. However, detecting changes at this level remains a technical challenge. My laboratory has developed a new technology termed the DGS (Ditag Genome Scanning) that aims to analyze the genome structure at kilobase resolution. We have performed extensive computational analysis and experimental evaluation of the technology. Our study indicates that the DGS technology provides a powerful tool for studying human genome structure.
On Mass Spectrometry Based Proteomic Data Processing Using Multiscale Tools
Don Hong, PhD; Professor
Department of Mathematical Sciences, Middle Tennesse UniversityFriday, February 8th, 2008 – 2:00pm
COOK 3118 A&B
ABSTRACT: Modern technology today allows the collection of biological information at an unprecedented level of detail and in increasingly vast quantities. To reap real knowledge from the mountains of data produced, however, requires interdisciplinary skills--a background not only in biology but also in mathematical tools and statistical techniques of data analysis. Mass spectrometry (MS) has become one of the critical components in medical research. The matrix-assisted laser desorption ionization (MALDI) technique allows the use of MS in applications involving large molecules. In 2002, the Nobel Prize in Chemistry recognized MALDI's ability to analyze intact biological macromolecules. Though MALDI MS has proven to play a key role in the advancement of science with the introduction of new fields such as Proteomics, there are many challenges both in MS data preprocessing and data analysis. In this talk, I'd like to present some recent progress on MS based proteomic data processing using multiscale tools such as wavelets and empirical mode decomposition (EMD) methods.
Computational Methods for Assessing Existing and Proposed Pharmaceuticals
Philip Bourne, PhD – Professor
Department of Pharmacology, University of California San DiegoFriday, January 18th, 2008 – 2:00pm
COOK 3118 A&B
ABSTRACT: We have recently developed a computational method to determine the geometric potential [1] which describes the ligand binding sites of 3-dimensional protein structures. Subsequently we developed a fast approach to search for these sites in a high throughput mode [2] across the druggable proteome. The goal of the project is to use these tools to search for competitive binding of major pharmaceuticals which might explain observed side effects, reposition an existing drug, or ultimately point the way towards further lead optimization. So far we have been able to offer an explanation for the side effects observed using select estrogen receptor modulators (SERMS), for example tamoxifen [3] and potentially reposition an existing drug for use in the treatment of drug resistant TB. Work with other drugs as well as some evolutionary implications of shared binding sites will be reported.
[1] L. Xie and P.E. Bourne 2007 A Robust and Efficient Algorithm for the Shape Description of Protein Structures and Its Application in Predicting Ligand Binding Sites BMC Bioinformatics, 8(Suppl 4):S9 [2] L. Xie and P.E. Bourne 2007 Detecting Evolutionary Linkages Across Fold and Functional Space with Sequence Order Independent Profile-profile Alignments. PNAS, Submitted [3] L. Xie, J. Wang and P.E.Bourne 2007 In Silico Elucidation of the Molecular Mechanism Defining the Adverse Effect of Selective Esterogen Receptor Modulators. PLoS Comp. Biol., 3(11) e217
Predicting Genes in the Dictyostelium discoideum and the Dictyostelium purpureum Genomes
Daniel Schneider – CBB Master's Student
Friday, January 25th, 2008 – 2:00pmCOOK 3118 A&B
ABSTRACT: The recent sequence and assembly of the Dictyostelium discoideum genome has made it possible to scan the entire genome with gene prediction software. We have chosen to use the Geneid software package along with a new parameter file created using the human curated Dictyostelium discoideum gene models. The new parameter file was based on nearly 2800 human reviewed gene models, whereas the initial gene prediction on the Dictyostelium discoideum genome relied on a parameter set generated using a distantly related species. Out of 13644 re-predictions, 671 were new, 10042 were identical and 2951 differed either by start site, end site, or exon structure. We have also utilized the Dictyostelium discoideum parameter file to predict 14,097 genes within the Dictyostelium purpureum genome.
Research Advisors:
Rex Chisholm, Feinberg School of Medicine
Warren Kibbe, Feinberg School of Medicine
Large-Scale Neural Connectivity Analysis using Graph Theory
Devdutta Warhadpande – CBB Master's Student
Friday, January 11th, 2008 – 2:00pmCOOK 3118 A&B
ABSTRACT: Functional Magnetic Resonance Imaging (fMRI) is used to detect regional changes in blood flow within the brain. Traditional analyses use statistical methods to determine highly activated brain regions, however, none of these methods interactions between brain regions and, to date, have not been used for large-scale brain connectivity analyses. Eguiluz et al. tested and validated the use of pair-wise correlations to generate a functional network, revealing that the networks have scale-free properties. Using fMRI data from a study at the Speech Research Lab, pair-wise correlations between all permutations of two gray matter voxel (basic unit of measure) intensities in the entire brain are calculated for each condition in all subjects; the correlation coefficients represent the probability of a connection between corresponding voxels. Assuming a threshold value, voxels with coefficients above the threshold are assumed to be connected and are converted to "1"; all other values are converted to zero. The resulting binary connection matrix is used to calculate a "degree" value (the number of connections) for each voxel and the overall frequency of that value; hence a higher degree translates into more connections. Degree-distribution plots reveal a power law relationship characteristic of scale-free networks in graph theory. When the degree values for each pixel are mapped back to the brain, results confirm previous work by Marguilis et al. and further support the effectiveness of this method. Our analysis generates an undirected, unweighted graph. At present we are using correlation coefficients to weigh each connection to generate a completely connected, undirected, weighted graph. Creating a weighted graph in this manner requires no assumption regarding the threshold value, and therefore makes the analysis more comprehensive.
Research Advisors:
Patrick Wong, Department of Communication Sciences & Disorders
JiPing Wang, Department of Statistics
David Chopp, Department of Engineering Science & Applied Mathematics
* Please note that this presentation has been re-scheduled from an earlier date
A Divide-and-conquer Strategy to Solve the Out-of-memory Problem of Processing Thousands of Affymetrix Microarrays
Chia-Ju Lee – CBB Master's Student
Friday, November 30th, 2007 – 2:00pmCOOK 3118 A&B
ABSTRACT: An extremely large amount of microarray data has been produced and accumulated. As of October 2007, 173,486 arrays have been deposited into the NCBI GEO database. It is anticipated that more information can be discovered from large-scale experiments or large compilations of experiments than from small experiments with just a handful arrays. However, processing such a huge volume of data requires tremendous computing resources, which is far beyond the capacity of most research labs. A frequently encountered issue is the out-of-memory problem when processing thousands of CEL files generated by the Affymetrix platform using Bioconductor. We propose a divide-and-conquer strategy to solve this problem. It works recursively by breaking down a problem into many sub-problems of the same type, until they become simple enough to be solved directly in the memory. The solutions to the sub-problems are then combined to give a solution to the original problem. We used the CAMDA 2007 META-analysis data set, which contains 5,896 microarrays, to test our approach. The results were validated against a gold standard data set obtained by using a main frame computer to run 5,896 arrays on a computer with 1TB of physical memory. In summary, this study is aimed at developing a general strategy to run any established Affymetrix pre-processing algorithms in the Bioconductor package on a commodity computer cluster (32-bit CPU and 1GB of memory for each CPU).
Research Advisor: Simon Lin, Feinberg School of Medicine
CBB Research Day - 2007/2008
CBB Research Day is a forum for faculty members to introduce CBB-related research projects to the CBB students. Research Day is a terrific opportunity for interaction and future collaboration between current CBB faculty members, students or faculty interested in joining our program and our current students.
Friday, November 16th, 20072:30pm - 5:30pm
TECH L324
RESEARCH PROJECT BOOKLET
Title and Abstract Forthcoming
Vanessa Wereko - IT Technical Lead
Abbott Laboratories
**Cancelled**
Friday, November 9th, 2007 - 2:00pmCOOK 3118 A&B
An Overview of Independent Component Analysis of fMRI Data
Omar Akbik – CBB Master's Student
Friday, October 12, 2007 – 2:00pmCOOK 3118 A&B
ABSTRACT: My research project is aimed at finding whether or not there are significant blood flow changes in the brain, specifically the visual and motor cortex, when acupuncture therapy is applied. We propose that through statistical methods to be discussed herein, conclusions can be made about the validity of the aforementioned statement. It was first necessary to understand basic acupuncture techniques, the basic principles of functional magnetic resonance imaging (fMRI) and its application and finally, independent component analysis (ICA) in order to carry out this research. This presentation will include a brief overview of the following items: acupuncture, fMRI, and ICA. I will then discuss the data that was obtained by Dr. Parrish and the programs used to manipulate the data, including the program used to perform the ICA. Finally, I will discuss the results of my analysis, problems encountered, and possible solutions.
Research Advisors:
Todd Parrish, Department of Radiology
JiPing Wang, Department of Statistics
Computational Analysis and Modeling of Brain Connectivity
Olaf Sporns - Associate Professor & Associate Department Chair
Department of Psychological and Brain Sciences, Indiana University
Friday, October 5th, 2007, 2:00pmCOOK 3118 A&B
ABSTRACT: The relationship between structure and function is of central importance for all biological systems, and it remains a particularly important challenge for our understanding of neural systems. My talk will be about emerging links between aspects of brain structure (connectivity) and brain function (dynamics and embodiment). Brain networks can be characterized by attributes such as reciprocal pathways, short path lengths, high clustering, an abundance of specific motifs, and highly economical wiring volume or length. How do these structural attributes relate to functional characteristics of brain networks, to their dynamic patterns, to their processing power, robustness, or capacity to support flexible behavior in embodied systems? I will review a series of computational approaches ranging from graph theory to robotics that attempt to identify how complex brain networks are organized, how they process and integrate information, and how brain, body and environment dynamically interact.
COMPLETE 2007/2008 SEMINAR SCHEDULE
SPRING 2007 SEMINARS