A custom collaboration with: Quintiles
Harnessing the combined power of data and technology promises to build a bridge from today’s medicine for the masses to tomorrow’s personalized treatments. In the already post-blockbuster era of pharmaceuticals, healthcare must—without question—erect and cross this bridge. Besides databases of genomes and biomarkers, electronic health records (EHRs) offer benefits to patients, physicians and even developers of new medicines—the latter, for example, by improving the capabilities of clinical trials and post-market surveillance of approved drugs. Although healthcare cannot construct this bridge in total today, existing tools supply a mounting list of useful building materials.
The pile of useful parts increased dramatically with the sequencing of the human genome in 2003. This sequence provided fundamental elements for turning more data than ever into resources to better understand biological mechanisms in health and disease. Even genomes of other organisms, such as parasites and viruses, can advance healthcare. Unraveling plant genomes might lead to improved medicines and vaccines. With so many genomes that could prove beneficial to healthcare, the amount of sequencing information expands by the day. In 1982, for instance, GenBank—a sequence database managed by the U.S. National Institutes of Health—contained data on 680,338 nucleotides and 606 sequences. By 2008, those numbers grew to nearly 100 billion nucleotides and almost 100 million sequences. Then, there’s Europe’s primary sequence database, EMBL-Bank, and Japan’s DNA Data Bank of Japan, or DDBJ. On one hand, this broad expanse of information provides a healthcare bonanza; on the other, it creates a raging torrent of data, one difficult to control and use.
Even in the face of the genomics-data mountain, other classes of information promise to grow even larger, and far more complex. Today’s useful data range from sequences and single nucleotide polymorphisms to molecular biomarkers and biochemical pathways related to specific diseases. Moreover, recent developments mark a turning point—according to Stephen Friend, president and co-founder of Seattle-based nonprofit Sage Bionetworks—“from viewing diseases as defined by symptoms to considering diseases as changes in the [biomolecular] network state of the person.” The network, he says, is made up of interacting proteins, and most diseases arise as a consequence of often subtle but profound shifts in a network component rather than something going haywire with a handful of genes, as in Huntington’s disease. “If we can think of the disease in terms of network models, then when we go to design a drug [and] biomarkers, we can frame that biomarker in the terms of the network,” Friend says.
Further still, with so much information, from gene sequences to biomolecular pathways, researchers must take charge of even more data, especially data on individual patients, to smooth the pathway to personalized medicine.
The amount of genetic data available, since the dawn of the biotechnology era, has vastly expanded from just 606 base pairs and 680,338 sequences in 1982. The biggest challenge for health technologies will be to process and make sense of the ever-growing banks of information.
© 2010 Scientific American,
a division of Nature America, Inc.
All Rights Reserved.