The complementary DNA (cDNA) probe is binding the two DNA strands to form a hybrid and each section in the tray has a different hybrid version.The cDNA probe is tagged with a molecular marker of either radioactive or fluorescent molecules, which can be detected with imaging techniques. RT/PCR stands for reverse transcription polymerase chain reaction.
CLICK ON IMAGE TO ENLARGE.

The use of high-performance computing (HPC) with new biochemisty, optical processing, and database technology is rapidly advancing the fields of genomics research, drug discovery, and personalized heathcare.

For example, new “hybrid” computing platforms that consist of fast microprocessors and integrated circuits known as field programmable gate arrays (FPGA) let researchers and scientists better amass and search huge amounts of genetic information to help pinpoint the origins of cancer and Alzheimer’s disease. The hybrid technology is driving innovation because it cuts costs while being faster than previous methods, such as Cray supercomputers or CPUs networked together in a cluster.

Bio-IT instruments are now capable of determining an individual’s genome, or entire genetic makeup. Here, third-generation devices determine a genome by “sequencing” millions of DNA bases. DNA is composed of 4 nucleotide bases: adenine, guanine, cytosine, and thymine, which are abbreviated as A, G, C, T.

With dye-terminator sequencing the nucleotide bases are labeled with fluorescent dyes, which emit light at different wavelengths and can be read when they are excited by a laser. Scientists look for similarities in regions between two DNA sequences to explore functional, structural, or evolutionary relationships.

Such genomic sequencing is being used in drug discovery to find key mutations in cancer and to identify patients who will respond to the drug. Genomic sequencing and analysis is much more efficient than traditional drug-discovery approaches involving large clinical trials. And FPGA’s accelerate parallel processing algorithms exponentially for gene sequence search, detection, and comparison in examining the alignment of DNA, RNA, or protein sequences. In bioinformatics, sequence alignment identifies regions of sequence similarity or difference. Mismatches can indicate mutations.

Designers can program an FPGA and customize its computation core and the internal datapath of the alignment algorithm for the needed number of bits. In one example, XtremeData in Schaumburg, IL used an FPGA accelerator to match 32 sequence tags against a 3 billion sequence gene and did so 2,000 times faster than a 2.2 GHz CPU. The accelerator runs the Smith Waterman (S-W) algorithm, a means of searching protein databases to find those with the best alignment, and is 200 times faster than a single core Opteron.

In the area of predictive medicine, the technology has bolstered the growth of “consumer genomics.” Easily administered and non-invasive genetics testing methods are already available. Just spit in a tube, send in a sample, and get back an extensive and comprehensive Web-accessed report of any genetic differences or “spelling errors” in the DNA being examined. Reports show an individual’s predisposition to cancer, heart disease, diabetes, and many other diseases.

Another growing application: pharmagenomics. Here, genomic information helps doctors select the optimal medicine and treatment for a person’s specific genetic makeup. This approach is being used to combat HIV, cancer, asthma, diabetes, and tuberculosis. As the costs of “personal” genomic sequencing continue drop, use of this approach will increase.

Needless to say, genetic sequencing results in large amounts of data and exponentially growing databases. The University of Buffalo Center for Computational Research is using an XtremeData dbX Analytical Data Warehouse Appliance, a system engineered specifically for exploring data sets more than 100 terabytes in size, to tackle such problems.

Continue on next page