Drug development is among the slowest, most failure-prone processes in modern science, with . Today, artificial intelligence methods have accelerated the first step 鈥� plucking promising molecules out of endless possibilities 鈥� but countless challenges remain. A successful drug must be not only safe and effective, but also able to bypass the body鈥檚 defenses and reach the right target.

Most drug candidates fail such optimizations. That鈥檚 where a new research institute at the 91爆料 has focused its attention. Housed in the 91爆料 School of Pharmacy, the brings together experts in artificial intelligence, drug discovery, pharmacology, data science and biotechnology to ease the bottleneck between promising molecules and successful drugs.聽

The Institute opened in July 2026 and is led by three 91爆料 faculty members: , an associate professor of medicinal chemistry who oversees the Institute鈥檚 AI-enabled molecular design; , the Milo Gibaldi Chair of Pharmaceutics and expert in drug metabolism and disposition; and , a professor of psychiatry and behavioral science in the 91爆料 School of Medicine, who leads drug discovery, translation and commercialization efforts.

91爆料 News spoke with the three co-directors about why drug candidates fail, how AI is speeding drug development and how I2D3 hopes to help get drugs to market more quickly.

What separates a promising molecule from a full-fledged drug? What properties need to be considered, and how can a developer work toward them?

Gaurav Bhardwaj: It really depends on the disease indication you are targeting and the therapeutic modality. Let鈥檚 say you have a promising molecule that interacts with the disease-causing protein. Delivery becomes equally important 鈥� do we need an orally delivered drug? Do we need to cross the blood-brain barrier? If the disease requires daily dosing, then injectable or IV methods aren鈥檛 optimal. If it鈥檚 delivered orally, then the molecule needs to be able to get across the gut barrier, and also needs to be stable enough that it doesn鈥檛 get chewed up by the body. It also needs to stay in the body for a reasonable time. A successful drug molecule has to meet all these and more criteria, and ultimately all these criteria are encoded by the sequence and structure of the molecule.

The Institute is devoted to aspects of drug development that are often overlooked. What problem do you see the Institute being able to help solve?聽

GB: Traditional drug discovery and development is a trial-and-error-based process. Either you find a useful molecule in nature and spend years optimizing it for human use, or you create many random combinations of molecules and hope that one of them has the function you need. Both of these approaches are highly unsuccessful, which has created a bottleneck.

Now the field is also focusing on an idea called rational drug design. It started long before AI but is now becoming even more common. People are using AI methods to design new molecules. However, a lot of that work has focused on the first step 鈥� finding a molecule that binds to a specific protein, or has a specific function in the body. That鈥檚 still not a drug, it鈥檚 just more candidates.

The bottleneck has now shifted. It鈥檚 no longer finding that first molecule, but now, how do you add all the other drug-like properties? That鈥檚 what the Institute is trying to do. Let鈥檚 build the models that ultimately make molecules that are going to be successful all the way through the drug development pipeline.聽

Marco Pravetoni: I see our work also as accelerating discovery. I work on substance use disorders, and my lab develops vaccines, antibodies and next-generation antibody-like molecules that target drugs in the body. With these new tools, instead of working to design 10 antibody candidates in a lab, we could design 1,000 or more, and then we can accumulate enough data to reduce any risks, so that what we bring to clinical trials is more likely to be successful. AI can do a lot of that.

How can you make it more likely that a drug candidate succeeds in trials?聽

Nina Isoherranen: Part of it is predicting what鈥檚 going to happen to a drug in humans before it鈥檚 ever given to humans. That should increase the success rate and eliminate the waste of doing a lot of unsuccessful trials.聽

We can also build machine learning and AI approaches to predict drug disposition in an individual person. What we talk about today are 鈥榙igital twins,鈥� which refers to a computational model of the individual patient and their characteristics. For example, how does your kidney function? What is your body mass index? And so forth. Then we generate a digital version of you. We can then predict how a certain drug would behave in your body and build the best strategy.聽

There鈥檚 also an access-to-treatment question here. Pregnancy is a great example 鈥� we often don鈥檛 know how drugs work in pregnant women because we鈥檝e never done trials. To be safe, we say that pregnant people shouldn鈥檛 take those drugs, but that means they don鈥檛 have access to a potentially hugely beneficial medication. If we can use AI and machine learning to predict how pregnant people respond to medications and how their bodies handle drugs differently from nonpregnant people we can make more medications accessible

Now with AI and machine learning, I think we can get to a place where we can truly sample the full space of possibilities.聽

How can the methods you鈥檙e building help with these individualized treatments?聽

NI: We know that drugs behave differently in different people. Even if we give them the exact same drugs and concentrations, people may still have different responses because of factors inherent to our bodies.

During drug development the candidate drug needs to be studied to see responses in different populations. Before you get a drug approved, you need to understand how liver disease, for example, is going to change exposure to that drug and whether you need to change the dosing. There鈥檚 a lot of guidance on drug interactions. Pharmacists manage drug interactions all the time, but it gets very complicated when you combine multiple patient factors. Now, if we have good predictive tools, we can predict what鈥檚 going to happen without having to do trials.聽

The ultimate goal here is to be able to predict, using model computational tools, what鈥檚 going to happen in individual humans before you ever give them a drug. What鈥檚 the right dose? The right timing?聽

91爆料 has established itself as a leader in these fields already. I鈥檓 thinking especially of the 91爆料 Medicine , whose director, , recently won the Nobel Prize in Chemistry. How does I2D3 fit into the broader 91爆料 ecosystem?聽

MP: IPD is a world leader in designing novel proteins, and the 91爆料 also has outstanding capabilities in clinical testing and implementation through the . However, there remains a critical translational space between discovery and clinical application 鈥斅� one that focuses on the pharmaceutical development needed to turn promising innovations into viable therapeutic products. That鈥檚 where I2D3 can play a leading role.

For example, when researchers at IPD develop a new protein, I2D3 can partner with them early to address formulation, manufacturability, stability, delivery, and other key pharmaceutical considerations that are essential for advancing a discovery toward the clinic and ultimately the marketplace. I2D3 would serve as a core translational partner, helping bridge the gap between innovation and implementation.

IPD brings unmatched strengths in protein design, ITHS provides expertise in clinical translation, and I2D3 contributes the drug development and pharmaceutical sciences capabilities needed to move discoveries across the translational continuum. Together, these organizations can create a powerful and highly integrated ecosystem.

For more information, visit . To reach the researchers, contact Alden Woods at acwoods@uw.edu.

For more than 20 years, the 91爆料 has been home to a database built, maintained and expanded around the goal of helping to prevent health complications from adverse drug reactions, one of the of injury and death in health care settings.

This year, the 91爆料 School of Pharmacy鈥檚 , or DIDB 鈥� the core research tool from the school鈥檚 nonprofit team 鈥� is celebrating both for innovation and two decades of independent funding through licensing agreements with companies, research institutes and regulatory agencies around the globe.

鈥淭he award from the American Society for Clinical Pharmacology & Therapeutics is a great acknowledgement of the impact we鈥檝e had in the drug development space,鈥� said Dr. , DIDB co-founder and director of Drug Interaction Solutions. 鈥淲e built something from scratch at the 91爆料, and now it is internationally recognized as an authoritative research tool, with over 180 organizations from 40 different countries as subscribers.鈥�

The Drug Interaction Database is a highly detailed, structured matrix of cross-linking entries designed to support research and regulatory scientists in academia, pharmaceutical companies and other organizations in their evaluation of drug interactions and drug safety. Entries for the database are curated by 91爆料 scientists from a wide range of drug-related documents, including clinical studies, drug developer publications, toxicity case reports and FDA New Drug Applications reviews.

The database is continuously updated as new information about drugs becomes available. Currently, the site has more than 170,000 entries involving in vitro (or “test-tube experiments”) and in vivo (in humans) data on metabolic enzymes and drug transporters (proteins in the body that help drugs pass from one organ to another); interactions with other drugs or with foods, herbs, tobacco and genetics; and other factors.

The Drug Interaction Solutions team of experts not only thoroughly reviews drug interaction information but also helps researchers use the system effectively.

鈥淭he impetus to initiate this database resulted from my work with antiepileptic drugs.

My eureka moment occurred in 1994 when I became able to segregate the clinical interactions of the drug phenytoin (Dilantin) according to two distinct but related enzymes. That 鈥榙iscovery鈥� propelled my efforts to pursue the development of the database,鈥� said , the founder of the DIDB and its principal investigator until 2009, when he retired from the 91爆料. Levy has in the field of drug disposition and drug-to-drug interactions聽and remains an advisor to the director.

鈥淭his award recognizes the excellence and dedication of the team of database researchers, as well as the input I received from colleagues in the departments of Pharmaceutics and Medicinal Chemistry in the School of Pharmacy, and the Department of Neurological Surgery in the School of Medicine,鈥� Levy said.

After establishing the plan for building the database and recruiting Ragueneau-Majlessi, Levy was able to gain funding initially through seed grants from several pharmaceutical companies. In 2002, the university began licensing access to the database through . Since then, Drug Interaction Solutions has remained a nonprofit venture with licensing revenues used to cover the costs of scientific and technical maintenance, as well as the development of new features.

鈥淭he DIDB is a prime example of a university-sourced innovation maintained by the university and made available as products directly to customers, as opposed to licensed to others or spun off as a company,鈥� said Ro茂 Eisenkot, senior innovation manager at CoMotion. 鈥淎s a longtime partner of the program, 91爆料 CoMotion has been collaborating with the team to build its licensing offerings and expand into new markets, while supporting all partner contracting activities such as risk management, managing distributors, fee collection and license renewals.鈥�

While the database is not intended for doctors in clinical settings to use directly, Ragueneau-Majlessi explained, it is evolving in that direction through the integration of its data into the tools that help doctors make drug choices and manage adverse drug interactions.

According to the FDA, two-thirds of patient visits result in a prescription, with more drug combinations being used to treat patients. Adverse drug reactions 鈥渋ncrease exponentially with four or more medications,鈥� the agency . In addition, herbals and food products (including fruit juices) can significantly affect various common medications, so multi-drug interactions are frequent in clinical situations.

鈥淚t should not be acceptable that a person can be given two drugs with a major adverse interaction when we know the mechanism behind that interaction,鈥� Ragueneau-Majlessi said. 鈥淲e have the mechanistic and quantitative understanding that allow us to predict drug interactions, and that is very powerful clinically. Adverse drug interactions can be prevented.鈥�

On its website, the Drug Interaction Solutions team the DIDB can support the growth of personalized medicine and the trend toward selecting the most appropriate drug and dose for each unique patient.

鈥淚 really believe that is the next stage for the database,鈥� Ragueneau-Majlessi said. 鈥淲e are now in the era of precision dosing and personalized therapy. And, even if we can鈥檛 prevent all drug interactions, we can manage them. If you understand the mechanism of the drug and its interactions, you can make sure that an individual patient is not negatively affected. Knowledge is power.鈥�

Levy and Ragueneau-Majlessi will officially receive the from the American Society for Clinical Pharmacology & Therapeutics in March at the society鈥檚 annual meeting. The award honors scientists in clinical pharmacology who 鈥渉ave demonstrated leadership in the application of significant, innovative science to clinical drug development.鈥�

###

For more information contact Marie-Christine Bodinier, senior marketing manager for Drug Interaction Solutions, at mariecb@uw.edu.