Self-taught AI system designs drugs from scratch

Self-taught AI system designs drugs from scratch

Researchers at University of North Carolina at Chapel Hill use an artificial intelligence approach that can teach itself to design new drug molecules from scratch. ReLeaSE stands for Reinforcement Learning for Structural Evolution, and its creators say they believe this application of machine learning can dramatically accelerate the design of new drug candidates.

ReLeaSE is an algorithm and computer program developed at the UNC Eshelman School of Pharmacy. It comprises two neural networks: a teacher and a student. The teacher knows the syntax and linguistic rules behind the vocabulary of chemical structures for about 1.7 million known biologically active molecules, said K. H. Lee Distinguished Professor Alexander Tropsha, Ph.D., one of the creators of the new AI system. “After learning the molecular alphabet and the rules of the language, the student starts creating new ‘words’, or molecules,” Tropsha said. “If the new word-molecule is realistic and has the desired meaning, the teacher approves. If not, the teacher disapproves, forcing the student to avoid bad words and create good ones.”

By working with the teacher, the student gets better and better at proposing molecules that are likely to be useful as new medicines.

Tropsha and Research Assistant Professor Olexandr Isayev, Ph.D., are the corresponding authors on the study. The lead author is Mariya Popova, a graduate student the School’s Division of Chemical Biology and Medicinal Chemistry and Laboratory for Molecular Modeling.

Isayev said that ReLeaSE is a powerful innovation to virtual screening, the computational method widely used by the pharmaceutical industry to identify viable drug candidates. Virtual screening allows scientists to evaluate existing large chemical libraries, but the method only works for known chemicals. ReLeASE has the unique ability to create and evaluate new molecules. “A scientist using virtual screening is like a customer ordering in a restaurant; what can be had is usually limited by the menu,” Isayev said. “We want to give scientists a grocery store and a personal chef who can create any dish they want.”

The UNC-Chapel Hill researchers have been able to use ReLeaSE to generate molecules with properties that they specified, such as desired bioactivity and safety profiles. “We have used the ReLeaSE method to design chemical libraries of molecules with many different properties,” Popova said. “We were able to customize physical properties, such as melting point and solubility in water, and more relevant to drug discovery, design new compounds with inhibitory activity against Janus protein kinase 2, an enzyme that is associated with leukemia.”

Tropsha said he believes that ReLeaSE should be of great interest to the pharmaceutical industry. “The ability of the algorithm to design new, and therefore immediately patentable, chemical entities with specific biological activities and optimal safety profile should be highly attractive to an industry that is constantly searching for new approaches to shorten the time it takes to bring a new drug candidate to clinical trials,” he said.

Subscribe to our newsletter

Related articles

Using machine learning to estimate COVID-19’s seasonal cycle

Using machine learning to estimate COVID-19’s seasonal cycle

Scientists are launching a project to apply machine learning methods to assess the role of climate variables in disease transmission

Brain-computer interfaces: AI reduces need for recalibration

Brain-computer interfaces: AI reduces need for recalibration

Machine learning will drastically improve brain-computer interfaces and their ability to remain stabilized during use, greatly reducing or potentially eliminating the need to recalibrate these devices.

4D printing may speed development of therapeutics

4D printing may speed development of therapeutics

Researchers have created a 4D printer capable of constructing patterned surfaces that recreate the complexity of cell surfaces.

AI challenge aims to improve mammography accuracy

AI challenge aims to improve mammography accuracy

AI techniques, used in combination with the evaluation of expert radiologists, improve the accuracy in detecting cancer using mammograms.

Using tumor-on-a-chip to find cancer drugs

Using tumor-on-a-chip to find cancer drugs

Researchers have developed a 'tumor-on-a-chip' that can better mimic the environment inside the body, paving the way for improved screening of potential cancer fighting drugs.

Childbirth: Visualizing the pain away with VR

Childbirth: Visualizing the pain away with VR

Could VR unlock a new frontier of pain relief for drug-free birthing? A PHD student is researching the tech's potential for pain management, so far the results are good.

Turbo chip for drug development

Turbo chip for drug development

Scientists develop process that facilitates and accelerates chemical synthesis and biological screening by combining all steps on a chip.

Arthritis: AI eliminates unnecessary treatments for children

Arthritis: AI eliminates unnecessary treatments for children

A machine learning algorithm was able to sort children with arthritis into distinct categories based on their patterns of inflamed joints in the body in a way that was also predictive of disease outcome.

AI identifies likely wildlife hosts for emerging flaviviruses

AI identifies likely wildlife hosts for emerging flaviviruses

After collecting data and comparing it with every known mammal and bird species on Earth, scientists have identified wildlife species that are the most likely to host flaviviruses such as Zika, West Nile, dengue and yellow fever.

Popular articles