Press "Enter" to skip to content

Can Artificial Intelligence Help Drug Development?

Copyright © DrugPatentWatch. Originally published at Can Artificial Intelligence Help Drug Development?

By Larry Ramer, DrugPatentWatch writer

Artificial intelligence, or AI, has advanced significantly in recent years and could significantly lower pharma companies’ R&D costs by enabling precision medicine and greatly reducing the amount of time needed to examine data on drug candidates. A number of small companies say that their AI technologies meaningfully streamline the drug development process.

AI Can Enable Precision Medicine

Artificial intelligence has become a much more effective because tech researchers have developed algorithms that enable computers to learn from events and patterns. Using this ability, known as machine learning, computers can examine huge amounts of data, such as the genomic information of millions of people, and use it to draw conclusions about the health of individuals.

More specifically, computers can use AI to find within data correlations between individual characteristics such as genes and lifestyle and the efficacy of various types of treatments and treatment dosages. Doctors can use these correlations to draw conclusions about the best treatment options for individual patients. This method of individualizing disease treatment is known as precision medicine.  Partly because of the increased certainty about drug efficacy that precision medicine enables, the phenomenon could reduce the amount of time needed to bring a drug to market and significantly lower pharma companies’ R&D costs.

Using AI for Drug Development

AI can also reduce drug makers’ R&D costs in a more direct manner.  Specifically, the technology could enable drug makers to select and evaluate drug candidates much more cheaply and quickly than is possible currently. Using existing methods, it takes an average of 12 years and more than $2.5 billion to create a drug and get it approved by the FDA. One of the main reasons that the process is so expensive and time-consuming is that searching through the existing data about the many thousands of available compounds is quite cumbersome and labor intensive. By quickly examining the data being collected around the world and using learned, pattern-based criteria to select compounds that are likely to be effective in treating diseases, AI systems can significantly streamline the process.

Trying to accomplish that goal, multiple small companies are using AI to analyze compounds and predict which of them have the best chance of successfully treating illnesses. For example, Britain’s BenevolentAI is utilizing the technology to “analyze vast quantities of bioscience information” and identify compounds that are likely to effectively combat specific diseases without causing serious adverse effects, Nvidia noted. The British company has acquired the rights to a number of clinical drug candidates and related patents.  After the company used artificial intelligence to evaluate the drug candidates, it believes that there is a good chance that they can be developed into “new medicines for hard to treat diseases “Johnson & Johnson’s Janssen unit signed a collaboration deal with BenevolentAI in November, 2016.

The CEO of AI developer Exscientia, Andrew Hopkins, who used to work at Pfizer, says that its technology can identify “drug candidates in roughly one-quarter of the time and at one-quarter of the cost of traditional approaches,” Reuters reported.

Exscientia, which reports that its “knowledge-driven systems design millions of novel, project-specific compounds and pre-assess each for predicted potency, selectivity, ADME and other key criteria,” signed a collaboration deal with GSK in 2017. Under the deal,  Exscientia’s AI enabled platform will be utilized in an effort to identify small molecules that can be used to treat up to ten targets selected by GSK.  The ten targets will span “multiple therapeutic areas,” the companies stated.  Exscientia has also signed a separate deal focused on metabolic diseases  with Sanofi..

A third company, Atomwise, says that its AI technology, called AtomNet, “helps predict the effectiveness of new drugs more rapidly” than conventional methods.  According to Atomwise, AtomNet researches small molecules with “unprecedented speed, accuracy and diversity.” The company has reportedly partnered with Merck on malaria treatments

Summing Up
Advances in AI could facilitate precision medicine and meaningfully speed up the drug R&D process, greatly reducing the amount of time and money needed to develop new drugs. Multiple small companies are attempting to use AI to make the drug development process faster and cheaper. Major pharma companies appear to believe in the promise of AI, as a number of them have recently signed deals with AI tech firms.

Make Better Decisions with DrugPatentWatch

» Start Your Free Trial Today «

Copyright © DrugPatentWatch. Originally published at Can Artificial Intelligence Help Drug Development?
DrugPatentWatch - Make Better Decisions