Over the last 20 – 30 years, pharmaceutical companies have heralded new breakthroughs in the discovery of new drugs. All of these were based on the promises made by modern advances in the computing capabilities of those times.
The first of these techniques was computer aided drug discovery (CADD). This relied on complex modeling techniques and high end graphical workstations that opened up a new field called ‘rational drug design’. While this did result in some important drugs, even including the early treatment for AIDS, it did not revolutionize drug research.
A few years later came a new technique called high throughput screening (HTS). Once again medical science thought it was on the verge of a disruption in discovery techniques of new medicines but the same story repeated. There were some discoveries, but not enough to justify the huge amounts of outlay required.
Part of the problem lay in the amounts of approximations that the researchers were forced to make. Drug discovery relied on modeling 3-D molecules of candidate drugs and creating 3-D models of receptor sites in the human body and then finding if there were possible matches that worked. This technique was so computationally heavy that it was just not possible to model tens of thousands of prospective drug molecules in a practical time frame. As a result, researchers worked on a large degree of approximation to get a reasonable degree of performance from computer systems of that period. Some simulations could last months even after simplifying the assumptions and making many compromises.
These compromises and approximations greatly affected the process of discovery. What worked in a computer model, often did not work in real life situations and the entire process had to be repeated. In the final analysis, this pushed up the cost of producing new drugs.
The Cloud Changes Drug Discovery
The entire scenario has changed with the advent of cloud computing. Rather than wait for 10 CPUs to work 100 hours to throw up a possible solution, it suddenly became possible to harness 1000 CPUs and get results within an hour. With this kind of performance improvement, the need for compromises and approximations vanished. The programs could be made as realistic and accurate as possible and yet the enormous computing power which was on call ensured that answers were forthcoming in a few hours at the most.
In one well publicized case, a large cluster was set up on a cloud computing platform, the work was done and the cluster shut down all in 20 minutes flat. And the cost? Essentially loose change at $6.40! In the traditional drug discovery world, it would have taken months to procure and set up the cluster itself and the costs would have been astronomical – all to be recovered from the end user.
Taking advantage of this capability, specialist software for drug design and discovery have begun to be available, these have a molecule database of more than 100 million compounds each with their individual physical properties, shapes, stereochemistry and chemical properties. There is no way this kind of data could have been used, had it not been for cloud computing services.
Yet another advantage of the liaison between drug discovery programs and cloud computing is that it had made experimentation and innovation possible. Previously, when a single simulation would run a month, every step had to be a deliberate, well thought out affair. There was just no room to follow a sudden train of thought or an inspiration. No ‘what if’ scenarios were possible. Today, however, with a simulation costing a few dollars and taking a few minutes, there is much greater experimentation going on. The results will be visible soon in new drugs and more radical treatment.
Finally, cloud computing has given the researcher the magic wand he always wanted. Maybe it will banish some deadly diseases and save some impoverished people by giving them access to cheap yet powerful drugs. This is the true disruptive power of cloud computing.
Be Part of Our Cloud Conversation
About the Guest Author:
Sanjay Srivastava has been active in computing infrastructure and has participated in major projects on cloud computing, networking, VoIP and in creation of applications running over distributed databases. Due to a military background, his focus has always been on stability and availability of infrastructure. Sanjay was the Director of Information Technology in a major enterprise and managed the transition from legacy software to fully networked operations using private cloud infrastructure. He now writes extensively on cloud computing and networking and is about to move to his farm in Central India where he plans to use cloud computing and modern technology to improve the lives of rural folk in India.