| 2005 Technology
Symposium > Biotechnology
Biotechnology
The Biotechnology TAT focuses on biomedical research as it intersects
with information technology, security, national intelligence, and defense.
This includes biomedical and neuroscience informatics, computational biology
and biologically inspired computation, biosecurity and biodefense, and
biosensing (including both sensing of biological agents and biologically-based
sensors).
BioComputation
Olivia Peters, Principal Investigator
Location(s): Washington and Bedford
^TOP
BioOntologies
Lynette Hirschman, Principal Investigator
Location(s): Washington and Bedford
^TOP
Biotechnology and Computational Biology
Jordan Feidler, Principal Investigator
Location(s): Washington
Problem
Biological agents present a significant challenge to homeland security and defense of the warfighter in asymmetric environments. The difficulty in dealing with biological threats is compounded by the relatively low barriers to entry to produce novel pathogenic agents. Improved techniques and methods combined with basic-level training can be exploited to aid in the design of new pathogens with increased virulence.
Objectives
This work will speed response to a novel pathogenic agent using computational modeling techniques to quickly identify how a biological agent acts to disrupt normal cellular processes. Our technical approach entails a process of iterative refinement whereby modeling and experimentation drive each other to increase our understanding of a particular cellular pathway commonly perturbed by biological warfare agents: Fas-mediated cell suicide.
Activities
Our initial focus is on creating a computational model of the Fas-mediated cell death pathway, which is disturbed by a number of biological warfare agents. We are working in collaboration with the Molecular Pathology Department at the Walter Reed Army Institute of Research, where we receive training on the experimental techniques that will be required to test the models.
Impact
Computational models will allow for rapid estimation of how pathogenic a novel agent may be so that countermeasures can be mounted that are commensurate with the posed threat. They will also shorten the time required to develop a possible pharmacological or antibiotic treatment by allowing researchers to explore alternative hypotheses in simulation and prioritize experimental approaches.
^TOP
Decision Support for Epidemic Control
Alfred Brandstein, Principal Investigator
Location(s): Washington
Problem
Mathematical modeling for decision support in biosecurity and infectious disease control has not yet matured. No method exists to determine systematically which information is needed and to choose the best course of action in the face of contradictory results from available models for evaluating the ramifications of decisions.
Objectives
This research seeks to help decision makers reach informed and sound decisions in the face of competing or conflicting options. Specifically, the project will develop a process and tools to compare and evaluate models for use in setting intervention strategy. We will also illustrate the use of agent-based modeling to identify high-risk locations in a transportation network.
Activities
We will study and reconstruct selected models that address epidemic intervention strategies and review common concepts in epidemiology. We will also construct, with the assistance of subject matter experts (SMEs), epidemic scenario models representing various disease occurrences and spreads by type, mechanism, and geographic area. The models will include parameters for known diseases but also will allow SMEs to include parameters for hypothetical diseases.
Impact
This research will establish MITRE's capability in epidemiological modeling and simulation, and position MITRE to assist decision makers in systematically evaluating strategies for epidemic control. It will also begin the development of a decision support system for biosecurity and control of infectious disease.
^TOP
Diagnosis of a Bio-Warfare Agent
Olivia Peters, Principal Investigator
Location(s): Washington
Problem
It is difficult to determine if a biological attack has taken place because the technology does not exist for rapid and early diagnosis of a biowarfare attack. Currently, we have no information infrastructure for efficient storage of relevant data, no method to identify the specific genes affected by the pathogens, and no algorithms for analysis and classification of unknown data.
Objectives
The objective is to develop a classifier to quickly identify if a warfighter has been exposed to a biowarfare agent, the agent used, and a timeframe for this exposure. We will do this by coordinating and managing a large amount of microarray data, performing feature extraction and dimensionality reduction, and designing a classifier to determine cellular pathogen exposure.
Activities
The focus in year one was developing feature selection and classification methods that could differentiate individual biological and chemical warfare agents from controls subjects. Year two has moved from a binary distinction to a multi class decision, classifying a sample by which agent it was exposed to.
Impact
The major impact will provide a rapid evaluation of exposure (agent and time). Additional impacts will include an ability to pinpoint the source of exposure, determine potential therapeutic techniques, and begin to understand unknown pathogens through their closeness to known agents. This project will add to and complement MITRE's existing biological expertise.
^TOP
Mathematical Modeling of Early Detection of Infectious Disease Outbreaks: Toward Real-Time Surveillance
Mojdeh Mohtashemi, Principal Investigator
Location(s): Washington and Bedford
Problem
Global health, threatened by emerging infectious diseases and bioterrorism, increasingly depends on the rapid acquisition, processing, and interpretation of massive amounts of data. Despite moderate advancements in data acquisition, real-time interpretation of data remains primitive. Early detection of infectious disease outbreaks requires timely and accurate detection of real-time epidemiological events for which current public health surveillance is inadequately prepared.
Objectives
We propose to develop mathematical and computational models for early detection of unusual epidemiologic trends based on historical and real-time data from collaborating hospitals and emergency departments, thus advancing the art of surveillance from post-epidemic detection to pre-epidemic detection. Such methods can be applied to a broad range of outbreaks of infectious diseases, whether naturally occurring or maliciously instigated.
Activities
We will acquire historical and real-time data from Harvard Medical School-affiliated hospitals and other collaborating hospitals. We will develop spatio-temporal and social contact structure models of early detection of infectious disease outbreaks. These models will be validated using expert assessments and standard statistical and simulation techniques, and incorporated into AEGIS, a real-time surveillance system at the Children's Hospital, Boston.
Impact
The project outcomes will provide the public health community with novel methods for early detection of infectious disease outbreaks, while advancing MITRE expertise in mathematical modeling of infectious disease and biosurveillance. This work will position us to play a critical role in public health surveillance and biodefense research and will support key MITRE sponsors, including DHS, HHS, DOD, and the IC.
^TOP
Neuroinformatics
Monica Carley-Spencer, Principal Investigator
Location(s): Washington
Problem
The neuroscience community is accumulating a vast amount of human brain mapping data that does not reach its full scientific potential because it is generally confined to the originating lab. While data may exist that a researcher could use to explore a hypothesis, the investigator may be unaware of it or does not have access to it.
Objectives
The overall goals of this research, conducted in conjunction with an external NIH grant, are to design, prototype, and evaluate an information infrastructure to help realize the full potential of a growing store of human brain mapping data. In this initial undertaking, we focus on a system that enables the analysis, exploration, and dissemination of structural magnetic resonance imaging data.
Activities
We have made significant progress toward developing image retrieval capabilities that will augment the NeuroServ data management platform developed under the NIH grant for managing and sharing neuroimagery. These capabilities include content-based image retrieval, or query-by-example, and image quality screening. Image features and metrics are tested with both synthesized and real MRI provided by collaborators in the NIH research community.
Impact
This project provides an important public service to the neuroscience research and clinical communities. But the problems facing these communities are not unique; they are isomorphic to those facing many of MITRE's traditional sponsors who must manage and exploit large quantities of imagery. We expect our research to transition readily to DoD, federal law enforcement and intelligence community spons
^TOP
Neuromorphic Framework
Mark Happel, Principal Investigator
Location(s): Washington
^TOP
Pathogen Capture Using Floating Films
Elaine Mullen, Principal Investigator
Location(s): Washington and Bedford
Problem
Recent events have heightened awareness of the need to protect drinking water against bioterrorist threats. Contaminated surface water can contribute to the spread of infectious disease through human and animal populations. To counter these threats, scientists need an inexpensive, unobtrusive method of concentrating and detecting harmful microbes and toxins in drinking water reservoirs and surface waters worldwide.
Objectives
We will design a prototype film to collect and concentrate specific pathogenic bacteria at the surface of water. We will optimize and quantify the film's stability, specificity, and efficiency under various environmental conditions and concentrations of organisms. During the course of our research, we will measure physical properties that could lead to the development of a remote water surveillance capability.
Activities
At Johns Hopkins Applied Physics Lab we will float designer films on water containing a mixture of pathogenic and harmless bacteria. A world-class team of experts will evaluate the experimental protocol and test results. We will measure the effects of varying micelle size and composition, pathogen concentration, and mixing time, and will periodically measure spectral characteristics of the films.
Impact
Films and micelles synthesized from lipids and glycoproteins offer an affordable and inconspicuous means of selectively concentrating pathogens at the surface of drinking water reservoirs. This technology may provide the basis of an affordable and unobtrusive large-area remote water surveillance system that could be licensed into commercial instrumentation. MITRE will establish valuable collaborative relationships with external labs and university teams.
^TOP
Patterns of Pathogenicity
Lynette Hirschman, Principal Investigator
Location(s): Bedford
Problem
New diseases are constantly emerging and concern with bioengineered weapons looms large in defense and public health planning. The rapid elucidation of the mechanisms (virulence factors) by which these microbes cause harm is key to an effective and rapid response.
Objectives
We propose to automate detection and classification of virulence factors based on analysis of the genome of a pathogenic organism. Biologists have identified classes of virulence factors; diverse pathogens share many of these virulence factors and often exchange "islands of pathogenicity." Using the genome to identify virulence factors is key to developing effective treatments, vaccines, and decontamination procedures.
Activities
We will bring the relevant pathogen data sets in house and apply pattern recognition, data mining, and computational biology techniques to identify features associated with possible virulence factors. We will develop interactive tools, working with Los Alamos National Laboratory (LANL). The LANL virulence factor database will be used to validate our detection and classification techniques.
Impact
This effort will provide new insights into the mechanisms of pathogenesis, while advancing MITRE's expertise in biology, bioinformatics, and biological threat reduction. We will partner with LANL to contribute to a key national database on virulence factors. This work will position us to play a major role in biodefense research, supporting key MITRE sponsors.
^TOP
|
