DNA Memory
Concept
Scientific datasets are extremely large, which makes analyzing, computing, and learning from the data extremely time consuming and resource intensive. PNNL has created techniques that reduce the burden of working with generated data, such as climate models, and observed data, such as that gathered by scientific instruments. We capture the essence of the large data set in a representation that is a fraction of the size. These smaller, visual representations of the data can be analyzed and compared with other data sets more quickly and easily.
Our researchers have developed a technique that could generate visual representations of sequential data such as data streams from brainwave or heartbeat monitors. We teamed with biologists interested in comparing and analyzing DNA data from genomes, which would typically fill pages and pages. Instead, we made a picture with the DNA data by assigning different colors to the nucleotides in DNA. By applying image-processing techniques, distinct features begin to emerge and areas for further study can be identified at a glance.