Life science research spans a wide range of varied topics beginning with our DNA (genomics) making our way to the cell (cell biology) and systems (developmental biology) all the way to disease research and pharmacology. Nowadays, conventional wet lab experimentation is being increasingly accompanied by mathematical and computational techniques, and these dry methods generate a staggering amount of data to analyze.
One the most important examples of the intersection of computation and biology is the advent of genomics which is the analysis of genome sequences and subsequently as -omics (genomics, proteomics and proteomics, etc.) that are leading the march toward more personalized medicine. However, rapidly advancing sequencing technology is both a blessing and a curse. On the one hand, as DNA sequencing costs virtually plummeting (http://www.genome.gov/sequencingcosts/), we now have access to cheap sequencing data, and lots of it. And yet, on the other hand, these advances in sequencing and the volumes of data they create are outpacing the advances in computing capacities, and consequently our ability to analyze the data. In our analysis, we will consider and extend hardware solutions being developed in companies such as Bina technologies that are speeding up the front-end of genomic processing by using customized off-the-shelf hardware consisting of high-end processors and very high-bandwidth memory. This brings the processing closer to the hospital with the goal of providing doctors data-driven suggestions for custom treatments.
Another important Life Science application, molecular dynamics is used for modeling protein docking which is a critical tool for drug discovery. Currently, it takes a long time to run molecular dynamics in standard Intel processors. However, we will also consider the hardware direction that the D.E. Shaw Group is following with respect to HPC and Big Data. They are using ASIC technology (Anton) to speed-up molecular dynamics for protein folding. They have already reduced the amount of time to obtain the data from days to minutes in hopes of decreasing the TTM for drug development.