Biomedical Engineering Reference
In-Depth Information
even with protective gear, are at risk for infection. A Biomolecular Database
query would be initiated to identify personnel who possess a known genetic
variation that prevents or mitigates infection. The personnel actually sent
into the contaminated area could then be selected from the list of genetically
resistant individuals.
As an alternative anti-terrorist application, suppose a large population (e.g.,
that of a city) has been exposed to a given biological or chemical agent. It then
becomes apparent that a subgroup of individuals require significantly more ag-
gressive medical therapy to survive, but for logistical reasons such aggressive
therapy cannot be provided to ALL exposed individuals. Stored DNA from re-
sistant and susceptible individuals can be used to determine the status of specific
groups of genetic markers as described in application
C
(markers are chosen
based on biological and medical inferences). In this way, a series of markers
diagnostic for increased susceptibility can be identified. This type of analysis is
called class discovery, and has been applied to the treatment of breast cancer and
leukemia, among other disorders. However, the use of Biomolecular Databases
can greatly streamline this work. Once diagnostic markers have been identified,
the techniques worked out in application 4.1.3 can identify individuals in need
of more aggressive care.
5.
DISCUSSION AND CONCLUSIONS
We have described Biomolecular Databases constructed from DNA for
rapid genetic analysis of large populations of individuals and complex diseases
involving multiple genetic loci. They may improve on conventional methods in
size of database and speed of search with the Biomolecular Databases system.
5.1. Comparison with Biomolecular Computing Methods for
SAT Problems
As described above, these selection operations can be executed by the use
of recombinant DNA operations, using logical processing methods developed in
the field of DNA computing. The methods used in DNA computing to solve
combinatorial search problems such as the Boolean satisfiability (SAT) problem
have the disadvantage that they require a volume that scales exponentially with
the size of the problem (i.e., the number of Boolean variables). This is because
the search space of possible Boolean variable assignments scales exponentially.
In contrast, logical queries are executed only on the information tags of the ex-
isting database, so the volume only scales linearly with the number of strands of
the Biomolecular Database.
