The mission of the Biomedical Informatics (BMI) training program is to provide graduate training in the application of information technologies to problems in biomedical research. The focus of the training is on the creation, validation and application of novel methods for capturing, representing, storing, retrieving, visualizing and analyzing biomedical data and knowledge. The Biomedical Informatics  program was founded in 1982 and broadly encompasses the areas of bioinformatics and clinical informatics.Trainees learn to work and communicate effectively at the intersection of contributing disciplines, including biology, medicine, computer science, probability and statistics, and the decision sciences. Trainees are expected to understand the ethical, legal and social implications of the technologies they use. Stanford provides an extraordinary environment to pursue interdisciplinary education in the development of novel informatics methodologies with applications spanning the full range of biomedicine.
    
Candidates in the BMI program may focus on research in any aspect of information management and analysis along the biomedical research pipeline. They are united in their interest in using information technology to manage, analyze and understand biomedical data, and in developing new approaches to using information to improve health care. Specific areas of investigation include: decision-support systems, knowledge acquisition, medical records, computational biology, systems biology, simulation, biological sequence analysis, biological 3D structure representation, pharmacogenetics, pharmacogenomics, genomics, collaborative technologies, network-based representation and retrieval of biomedical information and literature, medical imaging, reasoning under uncertainty, controlled terminologies for medicine and biology, technology assessment, and health-services research. The course of study requires training in the informatics methods used to represent knowledge and develop models, the computer science (CS) to implement these representations and models, and the specialized biomedical domain knowledge necessary to identify and make an impact upon important problems. Towards this end, students must take courses in 1) mathematics/CS that provide fundamental understanding of how knowledge is represented mathematically and how models are developed, 2) CS/biomedical informatics that develop understanding of how models are implemented and the technical requirements of the medium, e.g., programming languages, machine architectures, databases and algorithms, 3)  the bioscience curriculum that give them deep understanding of some area of biology or medicine, and 4) social policy and ethics that examine the societal impact of new technologies.

For more information contact:
Biomedical Informatics Training Program
Medical School Office Building (MSOB), room X-215
251 Campus Drive
Stanford, CA 94305-5479
650-723-1398 (phone)
650-725-7944 (fax)
bmi-contact@lists.stanford.edu
http://bmi.stanford.edu

To address diverse needs and backgrounds of students interested in biomedical informatics, the BMI program offers research-oriented graduate degrees and professional education. Candidates with training in computer science, the biosciences and other related fields are preferred.

 

Faculty and their Research Interests

Biomedical Informatics Core Faculty

Russ B. Altman. (Program Director). Research focuses on the creation of computational tools and resources to solve problems in biology and medicine.  Current projects are focused on three areas:  1) creating a database for how genetic variation in humans is associated with differences in drug response (pharmacogenomics), with particular recent emphasis on the drug warfarin, (http://www.pharmgkb.org/), 2) creating methods for identifying protein and RNA molecular function, in order to understand how we may engineer them to function differently (http://feature.stanford.edu/) and 3) understanding how physics-based simulation of biological structures can be facilitated at scales ranging from molecules to intact humans (http://simbios.stanford.edu/). Informatics methodologies include:  supervised and unsupervised machine learning, natural language processing, molecular dynamics simulations, database design, knowledge representation.

Atul Butte. Translational bioinformatics has been defined as the development of analytic, storage, and interpretive methods to optimize the transformation of increasingly voluminous genomic and biological data into diagnostics and therapeutics for the clinician. The research goal is to develop translational bioinformatics methods to reason over many available genome-scale measurement and experimental modalities, and apply these methods to study complex disorders in genomic medicine, especially obesity and type 2 diabetes mellitus.  The Butte Lab has four main directions in exploring integrative biology. First, we have developed bioinformatics methods to integrate genomic, genetic, phenotypic, clinical, and gene-knockout data from multiple sources and phenotypes and reason over these data. An example of this was our work in adipogenesis published in Nature Cell Biology (2005). Second, we have developed tools to automatically index and find genomic and proteomic data sets based on the phenotypic and contextual details of each experiment, published in Nature Methods (2007). We used these tools to create a comprehensive phenome-genome network published in Nature Biotechnology (2006). Third, we are building a novel gene-expression-based classification scheme for diseases across the entire field of medicine, as described in the New York Times and International Herald Tribune (2008). Fourth, we consider clinical measurements with molecular measurements to build multi-scale models of human health and disease, as published in Science (2008).

Amar K. Das. Research focuses on the advancement of computational technologies to support temporal reasoning, data integration, and collaborative systems in healthcare and the life sciences. Recent application areas involve research on autoimmune disorders and drug resistance.

Teri E. Klein. Research interests extend over the broad spectrum of pharmacogenetics, computational biology and bioinformatics. Applications include the development of a pharmcogenetics knowledge base, structure-function relationships, de novo modeling and the structural basis of disease.

Mark A. Musen. Research interests involve construction of automated systems to assist biomedical decision making, focusing on areas where the decision making is impeded by difficulties in formalizing knowledge and in encoding that knowledge for use by the computer. Current work addresses mechanisms by which computers can assist communities of scientists in the development of large, electronic knowledge bases.  The Protégé system (http://protege.stanford.edu) provides an experimental framework for investigation of collaborative knowledge-base development, of mapping among knowledge bases, and of knowledge-base visualization.  The National Center for Biomedical Ontology (http://bioontology.org) drives research on ontology-based access to biomedical data and knowledge, community-based peer review of electronic knowledge bases, and management of knowledge-base evolution.  The laboratory studies architectures for intelligent systems in areas as diverse as protocol-based therapy and surveillance for possible bioterrorism.

David Paik. Research interests lie at the intersection of radiology, molecular biology and informatics. We focus on developing and validating computational methodologies for extracting useful information content from anatomic, functional and molecular images, drawing upon image processing, computer vision, computer graphics, computational geometry, machine learning, biostatistics, modeling and simulation.

Daniel L. Rubin. Research interests focus on biomedical and translational imaging informatics.  We develop computational methods to identify and to extract information and meaning from images ("imaging phenotype") and to integrate and relate the image information to biological and clinical data ("molecular/clinical phenotype").  Our goal is to exploit images on a massive scale for discovery, similar to the data-driven approaches in modern bioinformatics, enabling us to discover image biomarkers of disease and to build predictive disease models from image data.  We translate our methods into practice by creating computer applications (such as decision support) that will improve diagnostic accuracy and clinical effectiveness.

 

Biomedical Informatics Affiliated Faculty

Serafim Batzoglou. Our lab is interested in the applications of mathematics and computer science to genomic research. Current research focuses on alignment algorithms, comparative genomics, gene regulation, regulatory motif finding, and microarray analysis.

Gill Bejerano. Our lab seeks to understand the human genome through vertebrate comparative, functional, and paleo-genomics, including such topics as the functional landscape of the human genome and its evolution, with particular focus on vertebrate gene regulation and its contributions to morphological diversity, development, and human disease; functions, origins, and evolution of proximal and distal cis-acting regulatory elements; the paradoxical existence of ultraconserved elements; co-option of mobile DNA elements (repeats) as a driving force in vertebrate evolution; and the interpretation of ancestral genomes.

Kwabena Boahen Our group has two synergistic goals: To understand how brains work; this will advance treatment of neurological diseases. And to build computers that work like brains; this will increase computational power a million fold. To these two ends, we are building large-scale neural models to link cellular-level biophysical processes with the system-level functions that they enable (e.g., cognition), through an interdisciplinary effort that brings electronics and computer science in contact with neurobiology and medicine.

Margaret Brandeau. Research focus is the application of mathematical andeconomic models to evaluate disease prevention and treatment programs. Current research focuses on HIV and drug abuse interventions, hepatitis B screening and vaccination, pandemic influenza preparedness, and bioterror response planning.

Douglas Brutlag. Professor Emeritus of Biochemistry and Medicine (by courtesy), teaches courses in Computational Molecular Biology (BIOC 218/BIOMEDIN 231),
Computational Genomics (BIOC 228/BIOMEDIN 228),  Computational Proteomics (BIOC 238/ BIOMEDIN 238) and Systems Biology (BIOC 278/ BIOE 310). The syllabi of some of these courses are located on the following websites: http://biochem218.stanford.edu/, http://biochem228.stanford.edu/ and 
http://biochem238.stanford.edu/

Markus Covert. Research focus is on building computational models of complex biological processes and using these models to guide an experimental program. Such an approach leads to a relatively rapid identification and validation of previously unknown components and interactions. Biological systems of interest include metabolic, regulatory, and signaling networks as well as intercellular interactions. Current research involves the dynamic behavior of NF-kappa B, an important family of transcription factors whose aberrant activity has been linked to oncogenesis, tumor progression, and resistance to chemotherapy.

David Dill. Our lab is interested in Boolean modeling to gaining insight into cellular processes at a systems level.  Our work includes analysis of Boolean circuit models using methods based on logic and automata theory, applied to understanding of the cell cycle, signal transduction networks, etc., and Boolean analysis of relationships in multiple large data sets, to understand regulation and global differences in gene expression among cell types.

James Ferrell. Research lab has two main goals: to understand the regulation of entry into and progression through mitosis and meiosis, and to understand the basic logic of signaling cascades and loops.

Sanjiv (Sam) Gambhir. Research focuses on merging advances in molecular biology with those in biomedical imaging to advance the new field of molecular imaging. Strategies for imaging cellular/molecular events in small animals and humans are in development. These include studying gene expression, signal transduction, enzyme levels and receptor levels in vivo. Use of these technologies for better management of cancer patients are emphasized. Mathematical modeling of these processes to better quantitate imaging data are also being pursued.

Alan M. Garber. Research is directed toward methods for improving health care delivery and financing, particularly for the elderly, in settings of limited resources.  He has developed methods for determining the cost-effectiveness of health interventions, and he studies ways to structure financial and organizational incentives to ensure that cost-effective care is delivered.  In addition, his research explores how clinical practice patterns and health care market characteristics influence technology adoption, health expenditures, and health outcomes in the United States and in other countries.

Mary K. Goldstein. Health services research in primary care and geriatrics. Ongoing work includes evaluation of methods of implementing clinical practice guidelines, for which she leads a multisite hypertension guidelines project using the ATHENA decision support system. Another research focus is evaluation of newly developed tools for automated guidelines, particularly for quality assessment.

Daphne Koller. Research focuses on applying machine learning and probabilistic methods to the analysis and reconstruction of cellular networks.  Current projects include the extraction of regulatory networks from gene expression data; the analysis of the effect of individual genetic variation on regulation and phenotype; and understanding how different network structures manifest in terms of gene expression, phenotype, genetic interactions, and more.

Jean-Claude Latombe. Research aims to create autonomous agents that sense, plan, and act in real and/or virtual worlds. Our work focuses on designing architectures and algorithms to represent, sense, plan, control, and render motions of physical objects. A key underlying issue is to efficiently capture the connectivity of configuration or state spaces that are both high-dimensional and geometrically complex. Applications include: robot-assisted medical surgery, integration of design and manufacturing, graphic animation of digital actors, study of molecular motions (folding, binding).

Michael Levitt. Research asks if it is possible to understand the molecular structure and function of proteins and nucleic acids in enough detail to make accurate predictions about structure and function. We are mounting a two-pronged attack on this problem using both molecular dynamics simulation and molecular modeling.

Henry Lowe. Primary research interests are in the areas of clinical and research information systems design, development and evaluation including multimedia clinical systems, integration of data to support patient care and clinical research, biomedical terminologies, automated indexing of biomedical documents, cancer information systems and biomedical data security.

Sandy Napel. Research focuses on CT and other medical imaging modalities. Our lab is currently interested in efficient and reproducible methods of extracting and visualizing medical information from the thousands of images typically generated by one or more radiological exams performed for each patient.

Douglas K. Owens. Research centers on four areas: technology assessment with a special emphasis on cost-effectiveness analysis, the application of decision theory to clinical and health policy problems, the development of methods for creation of practice guidelines, and decision support.

Dmitri Petrov. Three main topics are studied in the lab: 1) mutation and evolution of global genomic properties, 2) evolution and population dynamic of transposable elements in eukaryotes, and 3) evolution and population dynamic of transposable elements in eukaryotes.

Sylvia Plevritis. Research program focuses on computational modeling of cancer biology and cancer outcomes. We develop stochastic models of the natural history of cancer based on clinical research data. We predict population-level cancer outcomes under different screening and treatment interventions. We also analyze genomic and proteomic data in order to identify molecular networks that are perturbed in cancer initiation and progression and relate these perturbations to patient outcomes.

Gavin Sherlock. Research uses experimental laboratory and computational approaches to solve biological problems. We are using microarray technology to define all transcripts in the yeast genome, and to understand the changes in genome architecture and the transcriptome that occur in yeast as they evolve in vitro.  We are also developing novel yeast strains for use in biofuel production, with the aim of being able to ferment five-carbon sugars such as xylose, as well as 6-carbon sugars into ethanol.  In addition, we developed and run the Stanford Microarray Database, the Tuberculosis Database, the Candida Genome Database, and SOURCE.  Finally, we also write software for the analysis and visualization of microarray data, including GO::TermFinder, Caryoscope, and GeneXplorer.

Charles A. Taylor. Research focuses on the application of computational and advanced imaging methods to the study of the cardiovascular system. Applications of this research include the creation of knowledge and development of technology for the prevention, diagnosis and treatment of cardiovascular disease. Projects range from biologic research focused on disease processes to development of magnetic resonance imaging techniques to quantify blood flow and vessel strain to the development of new computational algorithms for simulating blood flow in human arteries.

Robert Tibshirani. Research is in applied statistics and biostatistics. Our lab specializes in computer-intensive methods for regression and classification, bootstrap, cross-validation and statistical inference, and signal and image analysis for medical diagnosis.

Michael G. Walker. Research interests include the genetics of disease, intelligence, and aging. He also provides statistics consulting to biotechnology and medical device companies and consults to venture capital companies evaluating investments in these areas.

 

Biomedical Informatics Collaborating Faculty

Stanley N. Cohen. The collection and interpretation of large amounts of data obtained from DNA and protein microarrays has become an important approach toward understanding the biological regulatory circuits that control gene expression. In the prevailing paradigm, clusters of genes that show common patterns of expression on microarrays are identified computationally and relationships among these genes are inferred by the experimenter in part by using his/her prior knowledge.

Ron Davis. Our lab is using Saccharomyces cerevisiae and Human to conduct whole genome analysis projects. The yeast genome sequence has approximately 6,000 genes. We have made a set of haploid and diploid strains (21,000) containing a complete deletion of each gene. In order to facilitate whole genome analysis each deletion is molecularly tagged with a unique 20-mer DNA sequence. This sequence acts as a molecular bar code and makes it easy to identify the presence of each deletion.

Stuart Kim. Research focuses on global analysis of conserved genetic modules. We are assembling all available DNA microarray data from several key organisms (human, mouse, fly, worm and yeast), and finding sets of orthologs that are co-expressed in multiple organisms. Research also focuses on molecular analysis of human aging: generating a molecular profile showing how nearly every gene is expressed in the kidney with respect to age.

Karla Kirkegaard. Our lab investigates the cell biology, genetics and biochemistry of RNA viral propagation, using poliovirus as a model system. For many subcellular viruses and parasites, RNA, not DNA, is the carrier of genetic information. Poliovirus serves as a model to increase our understanding of positive-strand RNA viruses for which no vaccine is available and which remain a significant health hazard: examples include other picornaviruses, such as rhinoviruses, coxsackieviruses and the deadly enterovirus 71, as well as more distantly related positive-strand RNA viruses such as hepatitis C and Dengue fever.

Garry Nolan.  The lab focuses on signaling in the immune system. Autoimmunity, cancer, leukemia and systems biology are prominent in our studies. We use Flow Cytometry (FACS) of phosphoprotein activation states in single cells in cancer and autoimmune disease (recent CELL and Science papers), machine learning of signaling status. We are using these techniques to study B and T cell signaling, dendritic cell function, and other immune parameters by analysis of biochemical functions at the single cell level.

Julie Theriot. Research concentrates on interactions between infectious bacteria and the human host cell actin cytoskeleton. Listeria monocytogenes and Shigella flexneri are unrelated food-borne bacterial pathogens that share a common mechanism of invasion and actin-dependent intercellular spread in epithelial cells. Our studies fall into three broad areas: the biochemical basis of actin-based motility by these bacteria, the biophysical mechanism of force generation, and the evolutionary origin of pathogenesis.

Paul (PJ) Utz. Research goal is to develop a better understanding of the pathogenic mechanisms underlying systemic lupus erythematosus (SLE) and other autoimmune diseases by exploring signaling pathways in blood cells, autoantibody production by B lymphocytes, and novel therapeutics targeting these and related pathways.