Biomedical researchers today increasingly employ a "translational" methodology wherein findings from various disciplines are translated into developing better diagnostics, therapeutic tools and healthcare policy. Ranging from running large scale computer simulations to assess the pharmacological response of drugs to creating virtual neurons mimicking aging, mathematicians and other computational types are now making key contributions towards understanding and remedying disease. The next logical step in this progression is the use of big data techniques to leverage the large amounts of clinical, genetic and epidemiological data that has become available. Some of the applications of big data are:

  • Augmenting traditional machine learning in systems biology. Uncovering networks of interacting biological entities is the primary goal in systems biology. This type of problem is already amenable to machine learning, a mainstay of big data predictive techniques.
  • Harnessing whole genome sequences. With newer developments in gene sequencing technologies, whole genome sequences for individuals will be commonplace. Thousands, and eventually millions of individual genomes on record in combination with associated data on disease and treatment histories is a typical problem suited to big data techniques.

Big data healthcareBig data is all set to play a central role in the clinical setting. Some of the accruals possible:

  • Evidence based medicine. The process of sifting through large data sets (both structured and unstructured) on clinical histories of millions of patients in order to predict the best and most up to date intervention or ideal combination of drugs for a particular condition. As clinical data evolves with the introduction of newer drugs or disease, so will the predictions based on big data techniques.
  • Treatment risk analysis. The ability to determine the likely risks and benefits associated with different drug combinations by exploring clinical data, which otherwise would not have been possible at the trial stage.
  • Personalized medicine. Given a patients clinical history and whole genome information (in the near future), the capability to personalize treatment and prevention.