Elementary, My Dear Watson…..
So said my Data Science guy when asked how anyone would stand up to IBM’s Watson’s firepower. We ask the right questions! Sounds very simple but it’s true. When there is a tsunami of data, it is not the quantity or even the quality of data that’s going to determine who will find something of value. It is someone who knows how to ask the right questions, which are both comprehensive and pointed at the same time. Talk about being a butcher and a surgeon, and still saving the patient!
“When you have eliminated the impossible, whatever remains, however improbable, is the truth.” That favorite aphorism by Holmes has come to haunt me these past few days, but as was true in fictional Victorian England, so it is true now in the new digital healthcare world. Nowhere is this more obvious as it is in discovering new ways to treat serious and complex diseases. One has to be comprehensive in the search for data and information that includes research material from sources such as, medical, clinical trial, basic sciences, public health….….Now, what happens if you ask a question that’s too specific? One ends up with the irritating “no results” response! But, if you are not specific enough, you will never reach the desired destination. So, what do you do? Much like the famous detective (or a trial lawyer, if that’s your favorite literary character), you build the case layer by layer, with each layer clarifying one part of the question. After building each layer of the pyramid, you finally realize that you are at the summit built on the foundation of a hypothesis well supported by iterative proof of concepts and substantiating evidence. The best example of this in medicine is the development of new targeted therapies in cancer, where a basic understanding of the cell cycle and differentiation have led to identification of molecular targets to treat cancer.
By understanding how normal cells work, one can work out what’s going wrong in cancerous cells that proliferate unrestrained. Significant discoveries have been made in the last decade by brilliant scientists resulting in a dozen Nobel prizes and even a couple of miracle drugs. So what’s new now? Well for one thing the Book of Life – the human genetic code has been written now. All of the seemingly disjointed molecular events from a myriad of experiments can be threaded together with the genetic code to yield the tapestry of life. All one needs is to have all of the molecular events and the power to thread the events in the right sequence to recreate the masterpiece of life.
This is the power of the Big Data approach that allows us to ask comprehensive questions in a sequential manner each yielding a reasoned answer building up to the epiphany of a breakthrough. Thus, one can start connecting genetic defects to molecular pathways and medical literature to unearth unrealized paths to the pot of new drugs at the end of the Big Data rainbow. So when you have access to all the data in the world (as anyone with an internet connection has), all one has to do is know how to ask the right questions. Add a dash of domain expertise in drug discovery and development to that and you have the promise of a thousand recipes for miracle drugs. So, when you have eliminated the impossible, you have to wake up to the promise of a thousand new truths each promising a breakthrough in your field of choice. So, (to end with another quote by Holmes) – the game is afoot!