The cost of developing a new pharmaceutical drug, from the research and development stage to market approval, runs at about $2.6 billion, according to a 2014 report published by the Tufts Center for the Study of Drug Development (CSDD) cited by the Scientific American. It also takes between 10 to 15 years.
Israeli scientists say they have developed a revolutionary smart method to discover and develop new drugs, based on artificial intelligence and machine learning, that will dramatically shorten preparation time and reduce costs.
Sign up for our free weekly newsletterSubscribe
Dr. Kira Radinsky, a renown data scientist and a visiting professor at the Technion – Israel Institute of Technology, and Shahar Harel, a PhD student at the university’s computer science department, presented their system late last month at the KDD 2018 conference in London, an annual event on Big Data and Machine Learning that draws prominent world academics and industry leaders.
Radinsky and Harel’s system seeks to tap into the modern-day, computerized processes of screening and selecting molecules with the greatest therapeutic potential – of which there are more than stars in the galaxy, making this an enormous task.
Their working hypothesis is that drug development “vocabulary” is similar to that of a natural language.
Harel said in a university statement that the system he and Radinsky developed, founded on artificial intelligence and deep learning, “acquired this language based on hundreds of thousands of molecules.”
“We are essentially presenting here an algorithm which addresses the creative stage of drug development – the molecule discovery stage,” said Harel. “This capacity leans upon our mathematical innovation, which enables the computer to understand the chemical language and to generate new molecules based upon a prototype.”
The researchers instructed the system to propose 1000 drugs based upon old drugs and were surprised to discover that 35 of the new drugs generated by the system are existing, FDA-approved drugs developed and approved after 1950.
Radinsky said in a statement that the system “is not only a means of streamlining existing methods but also entirely new drug development and scientific practice paradigms.”
“Instead of seeking out specific correlations based upon hypotheses we formulate, we allow the computer to identify these connections from within a massive sample size, without guidance. The computer is not smarter than man, but it can cope with huge amounts of data and find unexpected correlations,” she added.
Radinsky indicated that a similar computerized process is how, in another study, the scientists managed to find the unknown side effects of various drugs and drug combinations.
“This is a novel type of science which is not built upon hypotheses tested in an experiment, rather, upon data that generated the research hypothesis,” she said.
The Technion said in a statement that the breakthrough is particularly significant in light of Eroom’s Law, which asserts that the number of new drugs approved by the FDA should decline at a rate of approximately 50 percent every nine years. The term was coined in 2012 in an article published in Nature Reviews Drug Discovery and is a reverse order of Moore, the name of Gordon Moore, one of the founders of Intel. Moore observed that the number of transistors in a dense integrated circuit doubles every two years. In contrast, Eroom’s Law notes that each year, fewer and fewer drugs are marketed.
Dr. Radinksy projects that “this new development will accelerate and reduce costs of development of new and effective drugs, thereby shortening the time patients will have to wait for the drugs. In addition, this breakthrough is expected to lead to the development of drugs that would not have been generated with the conventional pharmacological paradigm.”
The system is currently being deployed for use in collaboration with pharmaceutical companies to further analyze the additional generated molecules, the scientists said.