Connect with us

The Conversation

Will AI revolutionize drug development? Researchers explain why it depends on how it’s used

Published

on

theconversation.com – Duxin Sun, Associate Dean for Research, Charles Walgreen Jr. Professor of Pharmacy and Pharmaceutical Sciences, University of Michigan – 2025-01-03 07:33:00

A high drug failure rate is more than just a pattern recognition problem.

Thom Leach/Science Photo Library via Getty Images

Duxin Sun, University of Michigan and Christian Macedonia, University of Michigan

The potential of using artificial intelligence in drug discovery and development has sparked both excitement and skepticism among scientists, investors and the general public.

“Artificial intelligence is taking over drug development,” claim some companies and researchers. Over the past few years, interest in using AI to design drugs and optimize clinical trials has driven a surge in research and investment. AI-driven platforms like AlphaFold, which won the 2024 Nobel Prize for its ability to predict the structure of proteins and design new ones, showcase AI’s potential to accelerate drug development.

AI in drug discovery is “nonsense,” warn some industry veterans. They urge that “AI’s potential to accelerate drug discovery needs a reality check,” as AI-generated drugs have yet to demonstrate an ability to address the 90% failure rate of new drugs in clinical trials. Unlike the success of AI in image analysis, its effect on drug development remains unclear.

Pharmacist searching through drawer of drug packages

Behind every drug in your pharmacy are many, many more that failed.

nortonrsx/iStock via Getty Images Plus

We have been following the use of AI in drug development in our work as a pharmaceutical scientist in both academia and the pharmaceutical industry and as a former program manager in the Defense Advanced Research Projects Agency, or DARPA. We argue that AI in drug development is not yet a game-changer, nor is it complete nonsense. AI is not a black box that can turn any idea into gold. Rather, we see it as a tool that, when used wisely and competently, could help address the root causes of drug failure and streamline the process.

Most work using AI in drug development intends to reduce the time and money it takes to bring one drug to market – currently 10 to 15 years and US$1 billion to $2 billion. But can AI truly revolutionize drug development and improve success rates?

AI in drug development

Researchers have applied AI and machine learning to every stage of the drug development process. This includes identifying targets in the body, screening potential candidates, designing drug molecules, predicting toxicity and selecting patients who might respond best to the drugs in clinical trials, among others.

Between 2010 and 2022, 20 AI-focused startups discovered 158 drug candidates, 15 of which advanced to clinical trials. Some of these drug candidates were able to complete preclinical testing in the lab and enter human trials in just 30 months, compared with the typical 3 to 6 years. This accomplishment demonstrates AI’s potential to accelerate drug development.

Drug development is a long and costly process.

On the other hand, while AI platforms may rapidly identify compounds that work on cells in a Petri dish or in animal models, the success of these candidates in clinical trials – where the majority of drug failures occur – remains highly uncertain.

Unlike other fields that have large, high-quality datasets available to train AI models, such as image analysis and language processing, the AI in drug development is constrained by small, low-quality datasets. It is difficult to generate drug-related datasets on cells, animals or humans for millions to billions of compounds. While AlphaFold is a breakthrough in predicting protein structures, how precise it can be for drug design remains uncertain. Minor changes to a drug’s structure can greatly affect its activity in the body and thus how effective it is in treating disease.

Survivorship bias

Like AI, past innovations in drug development like computer-aided drug design, the Human Genome Project and high-throughput screening have improved individual steps of the process in the past 40 years, yet drug failure rates haven’t improved.

Most AI researchers can tackle specific tasks in the drug development process when provided with high-quality data and particular questions to answer. But they are often unfamiliar with the full scope of drug development, reducing challenges into pattern recognition problems and refinement of individual steps of the process. Meanwhile, many scientists with expertise in drug development lack training in AI and machine learning. These communication barriers can hinder scientists from moving beyond the mechanics of current development processes and identifying the root causes of drug failures.

Current approaches to drug development, including those using AI, may have fallen into a survivorship bias trap, overly focusing on less critical aspects of the process while overlooking major problems that contribute most to failure. This is analogous to repairing damage to the wings of aircraft returning from the battle fields in World War II while neglecting the fatal vulnerabilities in engines or cockpits of the planes that never made it back. Researchers often overly focus on how to improve a drug’s individual properties rather than the root causes of failure.

Diagram of airplane with clusters of red dots on the wing tips, tail and cockpit areas

While returning planes might survive hits to the wings, those with damage to the engines or cockpits are less likely to make it back.

Martin Grandjean, McGeddon, US Air Force/Wikimedia Commons, CC BY-SA

The current drug development process operates like an assembly line, relying on a checkbox approach with extensive testing at each step of the process. While AI may be able to reduce the time and cost of the lab-based preclinical stages of this assembly line, it is unlikely to boost success rates in the more costly clinical stages that involve testing in people. The persistent 90% failure rate of drugs in clinical trials, despite 40 years of process improvements, underscores this limitation.

Addressing root causes

Drug failures in clinical trials are not solely due to how these studies are designed; selecting the wrong drug candidates to test in clinical trials is also a major factor. New AI-guided strategies could help address both of these challenges.

Currently, three interdependent factors drive most drug failures: dosage, safety and efficacy. Some drugs fail because they’re too toxic, or unsafe. Other drugs fail because they’re deemed ineffective, often because the dose can’t be increased any further without causing harm.

We and our colleagues propose a machine learning system to help select drug candidates by predicting dosage, safety and efficacy based on five previously overlooked features of drugs. Specifically, researchers could use AI models to determine how specifically and potently the drug binds to known and unknown targets, the level of these targets in the body, how concentrated the drug becomes in healthy and diseased tissues, and the drug’s structural properties.

These features of AI-generated drugs could be tested in what we call phase 0+ trials, using ultra-low doses in patients with severe and mild disease. This could help researchers identify optimal drugs while reducing the costs of the current “test-and-see” approach to clinical trials.

While AI alone might not revolutionize drug development, it can help address the root causes of why drugs fail and streamline the lengthy process to approval.The Conversation

Duxin Sun, Associate Dean for Research, Charles Walgreen Jr. Professor of Pharmacy and Pharmaceutical Sciences, University of Michigan and Christian Macedonia, Adjunct Professor in Pharmaceutical Sciences, University of Michigan

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More

The post Will AI revolutionize drug development? Researchers explain why it depends on how it’s used appeared first on theconversation.com

The Conversation

Can Trump just order new names for Denali and the Gulf of Mexico? A geographer explains who decides what goes on the map

Published

on

theconversation.com – Innisfree McKinnon, Associate Professor of Geography, University of Wisconsin-Stout – 2025-01-24 15:02:00

Known as Mount McKinley until 2015, Denali’s current name reflects what Native Alaskans call the mountain.
Arterra/Universal Images Group via Getty Images

Innisfree McKinnon, University of Wisconsin-Stout

President Donald Trump’s executive order to rename the Gulf of Mexico and Alaska’s Denali, the tallest peak in the country, has resulted in lots of discussion. While for some, such renaming might seem less important than the big problems the country faces, there is a formal process in the United States for renaming places, and that process is taken seriously.

Usually, so people don’t get confused, official, agreed-upon names are used by the government. In the U.S., place names are standardized by the U.S. Board on Geographic Names, which is part of the U.S. Geological Survey, the agency in charge of making maps.

In his executive order, Trump asks the Board on Geographic Names “to honor the contributions of visionary and patriotic Americans” and change its policies and procedures to reflect that.

Usually, renaming a place starts locally. The people in the state or county propose a name change and gather support. The process in each state is different.

A lake with sailboats and a city skyline in the background
Lake Bde Maka Ska, formerly Lake Calhoun, is the largest lake in Minneapolis.
YinYang/E+ via Getty

How to change a place name

Minnesota recently changed the name of a large lake in Minneapolis to Bde Maka Ska, which the Minneapolis Park Board described as “a Dakota name for the lake that has been passed down in oral history for many years.”

The board voted to change the name and took its request to the county commissioners. When the county agreed, the request was then sent to the Minnesota Department of Natural Resources, which made it official for Minnesota. Then, the state of Minnesota sent the request to the Board on Geographic Names, which made it official for the entire U.S.

It’s a lot of paperwork for something so seemingly minor, but people get passionate about place names. It took 40 years to rename Denali from the name established in the late 19th century, Mount McKinley.

The state of Alaska requested the name change in 1975, but the Board on Geographic Names didn’t take action. Members of the Ohio congressional delegation – President William McKinley was from Ohio – objected over many years to requests to rename the mountain, and the board did not act on those requests.

The president appoints the secretary of the Interior Department. The secretary works with the heads of related agencies to appoint the Board on Geographic Names. Current committee policy states, “Input from State geographic names authorities, land
management agencies, local governments, and Tribal Governments
are actively pursued.”

In 2015, President Barack Obama named a new leader for the Department of the Interior, Sally Jewell. Just as Obama made a trip to Alaska in late August 2015, Jewell declared the name change official under a law that allows the secretary of the Interior to change a name if the board doesn’t act on the proposal in a “reasonable” amount of time.

“This name change recognizes the sacred status of Denali to many Alaska Natives,” Jewell said. “The name Denali has been official for use by the State of Alaska since 1975, but even more importantly, the mountain has been known as Denali for generations. With our own sense of reverence for this place, we are officially renaming the mountain Denali in recognition of the traditions of Alaska Natives and the strong support of the people of Alaska.”

If someone objects to a name change, they could ask the courts to rule on whether the name change was made legally. Going back to Bde Maka Ska, some people objected to changing the name from Lake Calhoun, so they took the state natural resources agency to court. Eventually, the Minnesota Supreme Court ruled that the name change was done correctly.

Alaska’s two U.S. senators and prominent state figures have strongly objected to Trump’s renaming attempt.

How not to change a place name

Renaming the Gulf of Mexico is a different kind of case, however, from renaming a geographic place within U.S. borders.

The gulf is not within the territorial U.S. On the coast, the first 12 miles from shore are considered part of that country, but outside of that is international waters.

The Board on Geographic Names could change the name to Gulf of America on official U.S. maps, but there is no international board in charge of place names. Each country decides what to call places. And there is no official way for the U.S. to make other countries change the name.

It’s possible that the U.S. could formally ask other countries to change the name, or even impose sanctions against countries that don’t comply.

If the names were officially changed in the U.S., the government would use the new names in official documents, signage and maps. As for all the people and companies in the world that make maps, they usually use the official names. But there is nothing that would force them to, if they believed that a certain name is more widely recognized.The Conversation

Innisfree McKinnon, Associate Professor of Geography, University of Wisconsin-Stout

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More

The post Can Trump just order new names for Denali and the Gulf of Mexico? A geographer explains who decides what goes on the map appeared first on theconversation.com

Continue Reading

The Conversation

Trump inherits the Guantánamo prison, complete with 4 ‘forever prisoners’

Published

on

theconversation.com – Lisa Hajjar, Professor of Sociology, University of California, Santa Barbara – 2025-01-24 10:57:00

A control tower overlooks the Camp VI detention facility, at Guantánamo Bay Naval Base, Cuba.
AP Photo/Alex Brandon

Lisa Hajjar, University of California, Santa Barbara

President Joe Biden’s record of handling the U.S. military prison at Guantánamo Bay, Cuba, is decidedly mixed. He succeeded in reducing the detainee population he inherited by more than half, but he compounded problems in the military commissions that the Bush administration had invented in the wake of the 9/11 attacks to try people captured in the “war on terror.” Now all the problems at Guantánamo are again President Donald Trump’s.

When Biden took office in 2021, there were 40 prisoners. Today there are 15, the lowest number since the first 20 Muslim men and boys captured in Afghanistan were airlifted to the base on Jan. 11, 2002.

Biden left Trump four people the U.S. will not release but also cannot put on trial – the so-called “forever prisoners.” He also left intact the troubled military commissions system, with three pending criminal cases against a total of six detainees.

In December 2021, former chief military defense attorney Brig. Gen. John Baker testified before the Senate Judiciary Committee: “It is too late in the process for the current military commissions to do justice for anyone. The best that can be hoped for at this point … is to bring this sordid chapter of American history to an end.” Baker made clear that the only viable option is to resolve the cases with plea bargains for the defendants.

Marine Brig. Gen. John Baker tells U.S. senators that there is no opportunity for justice to be done at Guantánamo.

A chance to make progress

There are three cases that have not yet gone to trial – the 9/11 case with four defendants facing charges for their connections with the attacks, the USS Cole bombing in October 2000 with one defendant and the Bali bombing in October 2002 with one defendant.

The 9/11 and USS Cole cases have been stuck in the pretrial phase since Biden was Barack Obama’s vice president. In the summer of 2024, a breakthrough in the 9/11 case appeared imminent: Prosecutors and defense lawyers for three of the four defendants reportedly reached plea-bargain agreements. Khalid Sheikh Mohammad – the alleged “mastermind” of the attacks – Walid bin Attash and Mustafa Hawsawi agreed to plead guilty and accept life sentences in exchange for the government taking the death penalty off the table. There was no deal for the fourth 9/11 defendant, Ammar al-Baluchi.

The deals were approved on July 31 by the top military officer overseeing the Guantánamo commissions, retired Brig. Gen. Susan Escallier. But two days later, Biden’s defense secretary, Lloyd Austin, stepped into the process and overrode Escallier – whom he had appointed. Austin announced that the plea deals were revoked.

The judge, Air Force Col. Matthew McCall, decided to schedule plea hearings for early January. But after some legal back-and-forth that forced a stay, he had to cancel them. Biden left the case against three 9/11 defendants in limbo.

A beige one-story building with a sign above the door.
The basement of this government building in Bucharest, Romania, held a secret CIA prison, one of many across the world.
AP Photo

Witness to the transition

In mid-January 2025, I made my sixteenth reporting trip to Guantánamo. I came for closing arguments on a motion in the 9/11 case that seeks to suppress statements that Ammar al-Baluchi made to the FBI in January 2007. That was four months after he and 13 others were transferred to Guantánamo from CIA black sites where they were held for years. The litigation to suppress those statements started in 2019.

In Chapter 10 of my book, “The War in Court: Inside the Long Fight against Torture,” I detail how the litigation on this suppression motion made public previously unknown details and under-acknowledged horrors of the CIA’s rendition, detention and interrogation program.

These closing arguments were the culmination of six years of litigation on the key question in the 9/11 case: Does torture matter in the pursuit of justice in the military commissions?

A figure strapped to a table has water poured on its face.
A drawing by Guantánamo detainee Abu Zubaydah depicts a person being waterboarded.
Copyright Abu Zubaydah 2019. Licensed by Professor Mark Denbeaux, Seton Hall Law School

Can Guantánamo be closed?

Of the 780 people ever detained at Guantánamo, 540 were released during the presidency of George W. Bush, who established the detention facility. Obama, who signed an executive order on his second day in office pledging to close Guantánamo within a year, released 200.

In his first term, Trump pledged to keep the facility open. The only man to leave Guantánamo during Trump’s first term was Ahmed al-Darbi, who was repatriated to Saudi Arabia in 2018 to serve out the remainder of his sentence from a 2014 plea bargain agreement.

When Biden took office, he said that he supported shutting down the military prison at Guantánamo. In the early years of his presidency, there was a slow stream of transfers, mostly people who had been cleared for release long ago and were freed.

In Biden’s last months, the pace of transfers quickened. In December 2024, a Kenyan detainee, two Malaysian members of al-Qaida who had pled guilty the previous January, and a Tunisian man who had been in Guantánamo since the day the facility was opened were all repatriated to their countries of origin and freed. In January 2024, 11 Yemenis were transported from the prison to Oman to be resettled.

15 men left behind

The Biden administration had also planned to repatriate a severely disabled Iraqi detainee, Abd al-Hadi al-Iraqi, to serve out his plea-bargained sentence in a Baghdad prison. But a federal judge blocked that transfer, ruling that al-Iraqi would not get necessary medical treatment in Iraq and might be subject to abuse there.

Al-Iraqi is one of the 15 that Biden left behind. Three of them – a Libyan, a Somali and a stateless Rohingya – have long been cleared for release. Their continuing detention without charges highlights a key element of the Guantánamo problem: No one can be released unless the U.S. government finds another country willing to accept them.

One of the remaining detainees, Ali Bahlul, is serving a life sentence for conspiracy to commit war crimes. Six others, including the four 9/11 defendants, are awaiting their trials.

There are also four detainees whom the government refuses to transfer but cannot put on trial for lack of evidence.

A man wearing a beard, mustache and glasses.
The U.S. goverment says it cannot release Abu Zubaydah from Guantánamo because he would disclose classified interrogation techniques critics have labeled torture.
U.S. Central Command via AP

These so-called “forever prisoners” include Abu Zubaydah, a Saudi-born man of Palestinian descent who was taken into CIA custody in 2002 and was used as the guinea pig for the CIA torture program. The government long ago conceded that Abu Zubaydah was not a top leader of al-Qaida – in fact he was not even a member. But he will not be released because he knows how he was treated by the CIA, and that treatment remains highly classified.

The newest forever prisoner is one of the original 9/11 defendants, Ramzi bin al-Shibh; in September 2023, he was declared mentally incompetent to stand trial. Now he is uncharged, unreleased and untreated for his psychological maladies that were caused by the torture he endured in CIA black sites.

The ‘War on Terror’ is not over

When Biden pulled U.S. troops out of Afghanistan in August 2021, he claimed to have ended America’s longest war – and repeated this claim in a January 2025 speech. But the Guantánamo prison remains open, and as long as it is, the “war on terror,” which first put U.S. troops in Afghanistan in 2001, is not over.

How Trump will deal with Guantánamo is an open question. If he focuses on the death penalty, he will press ahead with military commission trials like his predecessors, hoping for unanimous guilty verdicts and death sentences. If he prioritizes cutting wasteful government spending, he will release additional detainees and allow the three plea bargain agreements to go into effect.

No one I spoke to during my last trip was willing to predict what a second Trump term might bode for Guantánamo – except that it won’t be closed.The Conversation

Lisa Hajjar, Professor of Sociology, University of California, Santa Barbara

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More

The post Trump inherits the Guantánamo prison, complete with 4 ‘forever prisoners’ appeared first on theconversation.com

Continue Reading

The Conversation

Red light therapy shows promise for pain relief, inflammation and skin conditions – but other claims might be hyped

Published

on

theconversation.com – Praveen Arany, Associate Professor of Oral Biology, University at Buffalo – 2025-01-24 07:43:00

A treatment typically lasts from three to 15 minutes.
Rich Legg/E+ via Getty Images

Praveen Arany, University at Buffalo

Red light therapy is increasingly viewed as a promising treatment for wrinkles, acne, psoriasis, scars and sun-damaged skin, and as a supportive therapy for some kinds of cancer. But does red light therapy live up to the hype that it’s practically a panacea for all sorts of ailments?

Praveen Arany is a professor of oral biology, biomedical engineering and surgery at the University of Buffalo and an expert on the uses of light and lasers for medical purposes. He explains how red light therapy works, for what diseases and conditions it may be most useful, and if red light home devices are effective.

What is red light therapy?

Treatment with red light therapy involves exposure to red light at a very low dose in a hospital or clinic.

It’s also called low-power laser therapy, soft laser therapy, cold laser therapy and nonthermal LED light therapy.

The umbrella term is called photobiomodulation therapy, which covers other colors, or wavelengths, that have health benefits. These light wavelengths span the visible to the near-infrared spectrum.

Red light is easily the most popular of the photobiomodulation therapies. That’s primarily due to its availability – the treatment has been around more than three decades.

While it’s true that other colors are also clinically and commercially available, researchers are still studying them to determine exactly how effective they are. That said, green light therapy is generally used to treat migraines; yellow light for depression; and blue light to kill resistant strains of bacteria, like MRSA infections, and to treat seasonal affective disorder, a depression that typically onsets in late fall and continues through winter.

The professional laser in the doctor’s office may be more effective than at-home LED devices.

How does red light therapy work?

Put simply, red light stimulates the cells in your body, energizing them while initiating blood flow to the affected area. That, in turn, spurs healing, similar to how your body responds to a cut by clotting the blood to heal a wound.

The treatment is simple and painless. The patient, either seated or lying comfortably, is exposed to the red light for three to 15 minutes. They may experience a feeling of warmth during treatment, but it should not be uncomfortable or hot. The clinician will likely recommend eye shields.

Used correctly, red light therapy is very safe. Overdosing – staying under the light too long or receiving treatments at very high power – does not necessarily cause harm, but it might reduce or slow benefits. However, just as some people are more prone to sunburn than others, some patients may be more sensitive to this light and might see redness in the skin. Those patients should receive lower light doses during treatment.

What medical conditions can the therapy help?

Randomized, controlled clinical trials show that red light therapy can reduce pain, inflammation and tissue damage. Because all of these things are prevalent in many illnesses, photobiomodulation may be a powerful adjunct for treating a wide range of diseases.

One example is cancer. There’s now strong evidence that red light therapy can lessen pain and inflammation from radiation, chemotherapy and bone marrow stem cell transplants. Red light therapy has also reduced other complications from cancer treatment, including oral ulcers, scars and fibrosis.

Other recent human clinical studies show that photobiomodulation helps heal diabetic and burn wounds, as well as some types of ulcers. However, this therapy should not replace good wound care treatment, such as disinfection. Photobiomodulation has also worked for patients with neck and back pain and tennis elbow.

What about other uses for red light therapy?

Although not proven to be effective by randomized controlled trials with large samples, which is the gold standard of research, red light therapy has been shown to benefit patients with Parkinson’s, Alzheimer’s, multiple sclerosis, fibromyalgia, arthritis, macular degeneration, myopia and autism in clinical case reports and lab research studies.

A word of caution, however: Red light therapy may not work for all the medical conditions that proponents say it does. Red light therapy is also used for cardiovascular health, elevating mood, relieving anxiety, improving muscle performance and recovery from sports injuries, and providing anti-aging benefits to the skin. While there’s some evidence to support these types of treatments, rigorous research studies are still missing.

Research indicates that red light therapy could help with myopia in children and macular degeneration.

What about its commercial use?

This is a rapidly evolving field. Both LED and laser devices – beds, lamps, helmets and face masks – are readily available in clinical and nonclinical settings, such as medical spas, gyms and beauty salons. They’re also available for at-home use.

Laser devices are more powerful and are typically found at a hospital, clinic or doctor’s office. An LED, or light-emitting diode, is less powerful and more often used in commercial or home settings.

The general consensus is that LEDs are OK to use in commercial establishments like beauty salons and medical spas, provided practitioners receive the appropriate training. But the use of laser devices should be relegated to clinical specialists. That’s because lasers, in untrained hands, have the potential to do more damage than LEDs.

As for some home products, their quality and reliability may be questionable; they might not meet minimum quality standards of output power or wavelength.

The U.S. Food and Drug Administration appears to be moving toward more rigorous evaluations of these products, especially with lasers, but there is a critical need for a certifying agency or body to take this on. These agencies would test the devices to make sure they’re actually meeting specifications. That hasn’t happened yet, but as it stands now, several scientific and professional organizations are exploring the possibilities.The Conversation

Praveen Arany, Associate Professor of Oral Biology, University at Buffalo

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More

The post Red light therapy shows promise for pain relief, inflammation and skin conditions – but other claims might be hyped appeared first on theconversation.com

Continue Reading

Trending