 |
 |
 |
 |
| |
|||||
|
|
|
|
|
|||||||||||
SOFTWARE-DEFINED BIOSENSING: By James Diggans SUMMARY: Current sensor technologies are relatively effective at detecting known biohazard threats but are slow to adapt in the face of new pathogens. To provide more flexibility and adaptability to biohazard responses, researchers are refining the concept of the software-based biosensor that can be updated quickly to confront emerging threats. Sensing in the Biological World The ability to sense the presence of deadly bacteria or viruses is critical to the missions of the military and public health and safety agencies. The distinction, however, between virulent pathogen and harmless organism is often minute. Modern biosensors not only must make that difficult distinction, but they must do it against a complex background of spores, mold, fungi, and pollen. Biosensors consist of two basic components: a capture element, responsible for detection, and a transducer, responsible for conveying the detection signal to a human being. All manner of increasingly exotic transduction methods exist—from fluorescent dyes to nanoscale cantilevers—and the pace of innovation shows no sign of slowing. Currently fielded biosensors, however, rely on capture elements designed to be specific to the organisms they seek to detect. For example, a biosensor designed to detect the causative agent of anthrax, Bacillus anthracis, may use protein-based antibodies or DNA-based probes that will bind only to anthrax to the exclusion of all other living organisms. The tradeoff for this exquisite selectivity is one of fragility in the face of change: a strain of B. anthracis could be intentionally engineered with subtle differences that would allow bacteria to escape capture without diluting its ability to cause disease. Towards Software-Defined Biosensing The capture challenge that biosensors face is not unlike that faced by our sense of smell. We live in an unpredictable universe of odors, all of which we would like to detect and classify. However, there is no way to decide on a single unchanging list of "important" odors and ignore the rest. Our noses have evolved to instead use many families of receptors that each bind to several odor molecules. The brain interprets the overall pattern of binding as a unique scent.
Biosensor researchers, including those at MITRE, are seeking to duplicate the basic strategy behind our nose's capture technique. Rather than relying on capture elements matched to individual pathogens, researchers are constructing a sensor that uses many non-specific capture elements. Software then analyzes the overall capture pattern to detect a defined organism. While editing small portions of a pathogen genome would be sufficient to evade existing sensors, to evade a software-defined biosensor would require wholesale editing of the genome while maintaining the microbe's capacity to cause disease. With current technology, this would be exceedingly difficult if not outright impossible. We Don't Know What We Don't Know An advantage of software-defined biosensing is that it provides agility in the face of an emerging and evolving biothreat. When a novel or engineered pathogen emerges, currently deployed biosensors must be physically updated. That is, a new capture reagent must be designed in a laboratory to be specific for the new threat. This capture reagent must then be mass-produced, which in the case of antibody-based capture reagents, is often difficult and time consuming. Once produced, these reagents must then be deployed to the sensors themselves. Many of the deployed sensors are capable of managing only a limited number of capture reagents. So in this case, tradeoffs must be made. Enabling the detection of the emerging threat comes at the cost of not being able to detect a previously known threat. By using non-specific capture elements paired with statistical models for detection and classification, software-defined biosensing—when faced with the threat of a novel or emerging infectious disease—requires only that updated models be sent to deployed sensors over existing reporting networks. It is no longer necessary to design a new capture element. You merely have to update the software to recognize a new pattern from its existing elements.
Detect to Treat ... or to Warn? Why this emphasis on agility in a deployed sensor network? For many known biological threats, treatments do exist—but to ensure that lives aren't lost, the treatments must be administered within hours or days. Current biosensors are designed to alert authorities of a potential biological attack in this space of time. Those exposed can then be located, brought to medical facilities, and treated. This warning approach is referred to as "Detect to Treat." However, with the current pace of biotechnology innovation, the ability to engineer novel pathogens will quickly outpace our ability to create new treatments for them. It took more than six months to produce the first dose of safe and effective vaccine for the H1N1, or "swine flu," influenza strain, even though we already understand influenza at the molecular level very well. For a pathogen with a novel mechanism of infectivity or toxicity, the amount of time needed to produce a treatment or vaccine would likely be much, much longer. For such unstudied threats, biosensing will need to move to a "Detect to Warn" approach. It will be crucial to provide warning within seconds to minutes of an attack to minimize the total number of casualties. In the face of this daunting timeframe, the ability to update a network of biosensors in real time with new capture capabilities will be a crucial step in mitigating the threat posed by a new pathogen. Because of its flexibility and adaptability, software-defined biosensing will offer the military and public health and safety agencies the timely warning they need to secure the health and safety of the population. Related Information Articles and News
Technical Papers and Presentations
Websites |
|
|||||||||||
For more information, please contact James Diggans using the employee directory. Page last updated: January 27, 2009 | Top of page |
||||||||||||
Solutions That Make a Difference.® |
|
|
