|
Informatics in Biology and Medicine |
 |
Open Faculty Position |
|
Institute for Genomics and Bioinformatics |
Research Areas
At UCI, our current areas of research in biomedical informatics include:
- Medical Information Access
- Knowledge Representation for Health-Care Guidelines
- Modeling Structure in Biomedical Data
- Biomedical Simulations
- Discovery of Gene Expression Control
- Knowledge Discovery in Clinical Databases
- Computational Biology
- Bioinformatics, Probabilistic Modeling and Machine Learning
Medical
Information Access
Patients and medical professionals
need easy access to vast quantities of medical information. The amount of
information in the primary medical literature alone is overwhelming. MEDLINE,
an on-line repository of medical abstracts, contains more than 8.6 million
bibliographic entries from over 3800 current biomedical journals and adds
31,000 new entries each month. To help people make effective use of this
information, researchers in medical information access are developing new
techniques to:
- automatically organize documents retrieved from a search
- visualize groups of documents among multiple dimmensions
- integrate information from a variety of sources
See also professor Wanda Pratt's pages.
Knowledge
Representation for Health-Care Guidelines
Current trends within the health-care
industry are for reducing cost and variability in health care practice.
This has led to a proliferation of national, regional and local guidelines
for health care. A significant challenge for medical informatics is the
representation and processing of complex guidelines. An important type of
guideline is the clinical trial protocol: a prescriptive guideline
for conducting experimental research in drug development or treatment procedures.
For these protocols, open research problems include:
- Enforcing standards of terminology, and aligning these with other existing medical terminologies
- Streamlining the protocol authoring process, and allowing for easy adaptation and modifications of protocols
- Evaluating protocol eligibility criteria, which could allow for automatic enrollment of patients into protocols. A simple decision support tool for enrollment into oncology clinical trials is currently being tested at the Chao Cancer Center.
- Tracking compliance with protocols, and providing decision support systems to help practitioners follow protocol recommendations
See also professor John Gennari's pages.
Modeling
Structure in Biomedical Data
Data sensing, acquisition and
storage technologies have led to vast observational data sets being routinely
collected in almost every aspect of biology and medicine. A significant
research challenge is developing theories, algorithms, and tools to handle
massive data sets so that scientists and clinicians can try to better understand
the phenomena generating the data. In particular, we are interested in developing
both predictive and descriptive models for structured data, including multivariate,
time series and sequences, spatial (2d and 3d), spatio-temporal, and longitudinal
(or repeated measures) data. Currently active projects in this area include:
- Mixture modeling of bivariate measurements of red-blood cell data for clustering and classification of subjects with iron deficiency (with Professor Christine McLaren, Dept of Epidemiology, UCI).
- Prediction of biological activity of molecular compounds based on high-dimensional sets of chemical features, for the purposes of improved drug disovery (with SmithKline Beecham Research).
- Segmentation of magnetic resonance images (MRIs) of subjects with Alzheimer's disease to relate volumetric changes in brain structure to clinical diagnoses of dementia and other measurements of cognitive function (with Dr. Pat Kesslak and Professor Carl Cotman, Brain Aging Institute, UCI).
See also Professor Padhraic Smyth's pages.
Biomedical Simulations
The MESSENGERS project is developing an infrastructure that permits processes, called Messengers, to migrate freely through a network of computers. One of the main applications of this paradigm are biomedical simulations. One type of such simulations include individual-based systems, where groups of individuals, each implemented as a Messenger, coexist in a simulated space and interact with one another. For example, a school of fish may be modeled by programming the behaviors of the individual fish, while the group behavior emerges automatically from interactions among the individuals. A second class of applications of the MESSENGERS system have been circulatory simulations, in particular in the areas of Toxicology and cardio-vascular modeling. In these applications the different organs of the human body are represented as nodes mapped on different computers. Messengers then mimic the flow of blood through the system by repeatedly visiting the different nodes and invoking computations to recompute the relevant values, such as changes in pressure, volume, or toxin concentrations, over time. For additional information,
see the MESSENGERS
project page.
The participating faculty members are
Lubomir Bic and Michael
Dillencourt
Discovery of Gene Expression Control
The simplified goal of the Human Genome project is to determine the entire sequence of the human genome. In itself, this goal is not useful. The genome reveals neither the function of genes nor the control of gene expression, which determines which functions are active. It is this knowledge that provides a scientific understanding for biological processes, and that, in turn, provides a theoretical grounding for medicine.This research project involves two professors from ICS and two professors from Biological Science. Our first goal is to find the regulatory elements, ie. determine the switches that control gene expression. The research involves a mixture of machine learning methods (classification and clustering) with biological evaluation. More distantly, we aim to define the network of regulatory elements that will enable the complete simulation of a cell's chemistry.
The participating ICS faculty
members are Dennis Kibler
and Richard Lathrop.
The Biological Science faculty members are Suzanne
Sandmeyer and
Calvin McLaughlin.
Knowledge Discovery in Clinical Databases
Michael Pazzani and Padhraic Smyth have been working in the area of knowledge-discovery for many years. One of the thrusts of their research has been the discovery of new knowledge in medical databases that would be useful for diagnosis and treatment. Clinical databases are particularly interesting since they contain a variety of heterogeneous information, included images, medical history, symptoms, and test results. In a collaboration with Professor Carl Cotman from the college of Medicine, the ICS researchers are applying advanced machine learning concepts as a tool to predict dementia in a large Alzheimer's disease database developed at UCI.
Bioinformatics, Probabilistic Modeling and Machine Learning
Our group works at the intersection of biological and computer sciences, using probabilistic/machine learning techniques to address biological problems and mine large data sets produced by massive data acquisition technologies, such as genome sequencing, high-throughput drug screening, and DNA microarrays. Current projects include the prediction of protein secondary and tertiary structure, the study of DNA structure in relation to several biological processes (protein binding, gene regulation, triplet repeat expansion diseases), and the analysis of gene expression data.
We also have a long-standing interest in more philosophical issues related to bioethics and what it means to be human in light of the current technological revolution in biology and computers, as exemplified by cloning, the Human Genome Project, and the Internet. This is an effort to foresee and recast progress in different areas of computational biology within a broader set of concerns.
See also professor Pierre Baldi's pages.
Computational
Biology
There are a number of significant
problems in biology and medicine for which computational approaches can
yield important insights. The world-wide efforts to construct databases
of protein and small molecule structures, DNA sequences, metabolic pathways,
regulatory mechanisms, pharmaceutical structures and activities, patient
response data, etc., have created many opportunities for intelligent systems.
Central questions include: What function is encoded in a protein sequence?
What structure will it fold into? How can we make better pharmaceutical
drugs? What factors effect patient response to treatment? Partial solutions
to these problems have been found using extensions of research on knowledge
representation, search, and learning. ICS faculty are involved in a project to develop a knowledge-based systems for recommending a customized multiple-drug therapy for HIV infected patients. We are also exploring an approach to creating knowledge-based systems by learning from patient data and have identified guidelines for screening for forms of dementia such as Alzheimer's disease. We have developed knowledge-based approaches to protein structure prediction, and implemented novel sequence-structure search algorithms and recognition methods. We are modeling DNA mutation and repair in connection with cancer-related studies.
See also professor Richard Lathrop's pages.
Faculty
Course Requirements
Core courses (all of the following)- ICS 208 – Introduction to Medical Informatics
- ICS 277A – Representations & Algorithms for Molecular Biology
- ICS 209 – Seminar in Informatics in Biology and Medicine (at least 3 quarters)
- ICS 205 – Human-Computer Interaction
- ICS 206 – Knowledge-based User Interfaces
- ICS 207 – Information Retrieval, Filtering, and Classification
- ICS 215 – Advances in Database Management Systems Technology
- ICS 227 – User Interfaces and Software Engineering
- ICS 234A – Computerization, Work, and Organizations
- ICS 266 – Computational Geometry
- ICS 273 – Machine Learning
- ICS 274 – Probabilistic Learning: Theory and Algorithms
- ICS 275B – Network-based Reasoning/Belief Networks
- ICS 276A – Neural Networks
- ICS 276C – Cognitive and Computational Neuroscience
- ICS 277B – Probabilistic Modeling of Biological Data
- ICS 278 – Data Mining
- ICS 280 – Special topics taught by one of the faculty in this area (only one 280 course counts toward the advanced topics requirement)
- Eng 210A – Systems, Anatomy, and Physiology I
- Soc Ecol 226 – Environmental Health Sciences III: Biostatistics and Epidemiology
- Bus 283 – Decision Analysis
- Ecology & Evolutionary Biology 251 – Molecular Evolutionary Methods
- Biological Chem 204 – Problems in Genomic Analysis
- Molecular Biology & Biochem 203 – Structure and Biosynthesis of Nucleic Acids
- Molecular Biology & Biochem 204 – Structure and Biosynthesis of Proteins
- Molecular Biology & Biochem 240 – Macromolecular Structure, Function, and Interaction
- Physiology & Biophysics 202 – Cellular and Molecular Neuroscience
1. Theory
- ICS 260 – Fundamentals of the Design and Analysis of Algorithms
- ICS 261 – Data Structures
- ICS 263 – Analysis of Algorithms
- ICS 212 – Embedded Systems Concepts
- ICS 241 – Computer Systems Architectures
- ICS 243 – Computer Networks
- ICS 252 – Introduction to Computer Design
- ICS 221 – Software Engineering
- ICS 211 – Compiler Construction
- ICS 247 – Distributed Computer Systems
- ICS 214 – Databases
- Bio Sci 94 – Diversity of Life
- Bio Sci 98 – Biochemistry
- Bio Sci 99 – Molecular Biology
- Bio Sci 108 – Developmental and Cell Biology
- Bio Sci 109 – Physiology
Paper Requirement for the Ph.D. Degree
Each student must write a survey paper and a research paper of publishable quality.
Related Links
Organizations:
- American
Medical Informatics Association (AMIA)
- International
Medical Informatics Association (IMIA)
- International
Society for Computational Biology
- Medical
Library Association (MLA)
- The
Medical Records Institute
- National Center for Biotechnology Information (NCBI)
- National
Library of Medicine (NLM)
- Society
for Medical Decision Making (SMDM)