Gillian Dohrn

The secrets of idiopathic pulmonary fibrosis (IPF) are written in its very name. Idiopathic refers to a disease of unknown cause, and the condition, which turns healthy lung tissue into fibrous scar tissue, still raises many questions.

IPF originates at the periphery of the lung and progresses inward, compromising more and more tissue and, eventually, making it difficult for a person to breathe. There is no cure for IPF, and neither of the two drugs that are approved as treatments can reverse the scarring—they only slow it down.

 researchers at Yale School of Medicine and collaborators took a significant step toward understanding IPF—and numerous other complex diseases—with an algorithm that interprets disease data and proposes treatments.

The research team developed a deep generative neural network called UNAGI (unified in-silico cellular dynamics and drug screening framework) that can identify patterns in disease-specific data. In a matter of hours or a couple of days, depending on the computer, UNAGI can learn how to glean insights from hundreds of thousands of cells, differentiating between cell types, picking out genes that are involved in disease progression, and identifying relevant regulatory networks. Then, it tries out different drugs, pulling from a long list of approved compounds to see if any of them work against whichever disease is being studied.

Although UNAGI was developed with IPF data, it can also be applied to changing physical states, such as aging, and other diseases, which the researchers demonstrated using COVID data.

“The model looks for regulation—what characterizes and regulates changes—and then, using known drug databases, also suggests treatments,” says co-senior author 

, Boehringer Ingelheim Pharmaceuticals, Inc. Professor of Medicine (Pulmonary) at Yale School of Medicine.

The research was done in collaboration with scientists at McGill University in Canada, KU Leuven in Belgium, and several other institutions.

In IPF, disease distribution is variable, with some areas becoming “sicker” than others. Several years ago, Kaminski and collaborators at KU Leuven developed a method for tracking IPF progression that doesn’t require repeated follow-up with the same patients—which can be a challenge for researchers—by grading disease progression within a single sample.

The KU Leuven team collected diseased lungs during transplant surgeries and cut them into slices, selecting small pieces to represent different stages of the disease. The Yale team cataloged gene expression patterns in individual cells from these samples, yielding the 

“We made some key discoveries, including novel cell types and populations,” says Kaminski. But the researchers wanted to understand disease trajectory better and lacked the analytical tools to do so with their data. “So, we started thinking about a way to apply AI to this problem.”

 an assistant professor at McGill University School of Medicine, who leads a lab that specializes in computational biology. The researchers developed UNAGI using sequencing data from 230,000 cells, provided by Kaminski.

“Our model is designed to represent virtual cells and virtual disease progression,” says Ding. Unlike most existing models, which are generic, UNAGI is disease-informed, meaning that it models the disease in question by identifying associated genes and regulatory networks. It then loops that information back into the model, refining its representation of cells and disease progression to add nuance. UNAGI also requires minimal researcher oversight, learning autonomously through an embedded iterative refinement process. By contrast, other models must be manually re-trained to interpret new datasets or test different drugs, which can be expensive and time-consuming.

UNAGI bypasses this need by integrating new information along the way, going deeper and deeper into the data until it can say with confidence which cells, genes, and pathways are involved in disease progression. “The model evolves to understand more and more about the disease,” says Ding. “It’s a bidirectional exchange of information.”

Then, pulling from a database of thousands of drugs with known mechanisms of action, UNAGI can test thousands of compounds and spit out a short list of potential therapeutics. In this study, it identified eight possible drugs, one of which is already used for IPF. The researchers selected one from the list that seemed out of place for their validation studies. Their pick—a calcium channel blocker called nifedipine—is used to treat hypertension, but UNAGI thought it might also have anti-fibrotic effects.

UNAGI is hitting pathways we did not think about before.
Naftali Kaminski, MD

When the researchers applied the drug to slices of human lung tissue that were designed to model IPF, the drug blocked the formation of scar tissue, as UNAGI predicted it would. Even if nifedipine doesn’t end up being a good fit for fibrosis, says Kaminski, “UNAGI is hitting pathways we did not think about before.”

Merging technologies like single-cell sequencing with AI will shape the future of this field, he adds. “We have a convergence of both very sophisticated AI-based analytical methods and the ability to generate data that is high-resolution enough to actually make these observations.”

Researchers at Hannover Medical School in Germany, Pacific Northwest National Laboratory, Baylor College of Medicine, and University of Pittsburgh also collaborated on this work.

The research reported in this news article was supported by the National Institutes of Health (awards R01HL127349, R01HL141852, U01HL145567, R21HL161723, P01HL11450, and U01HL148860) and Yale University. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. This work was also supported by the U.S. Department of Defense, the Else Kröner-Fresenius Foundation, CORE100Pilot (Advanced) Clinician Scientist Program of Hannover Medical School, and the German Research Foundation.

Articles

Idiopathic pulmonary fibrosis (IPF) is a progressive lung disease with no cure, characterized by fibrous scarring. Researchers at Yale School of Medicine developed UNAGI, a deep generative neural network, to analyze disease data and propose treatments. UNAGI identifies disease-specific patterns, differentiates cell types, and tests drugs from a database, suggesting potential therapeutics. In a study, UNAGI identified nifedipine, a calcium channel blocker, as a potential anti-fibrotic agent. This AI-driven approach, integrating single-cell sequencing, offers a promising direction for understanding and treating complex diseases like IPF.

Yale researchers developed an AI model that illuminates critical details about—and possible treatments for—idiopathic pulmonary fibrosis. It could do so for many other diseases.

AI Helps Researchers Understand Lung Disease and Proposes Treatment

Iron deficiency—when there’s too little iron in the blood—may affect a quarter of the world’s population, and in particular, women of reproductive age. Symptoms can include fatigue and shortness of breath, and without treatment, iron deficiency can progress to anemia, a condition characterized by low red blood cell count that can cause more severe heart and health problems.

There is a simple blood test for iron deficiency that measures ferritin—a protein in the blood that stores iron—but it’s not part of a standard checkup in most countries, including the United States. Further, the World Health Organization has, for decades, earmarked 15 micrograms per liter of ferritin as the threshold for low iron, which is too low, according to some experts, and does not reflect the true prevalence of iron deficiency. This lack of standardization creates blind spots, leaving many symptomatic individuals without access to treatment.

“We know for a fact that we’re missing people who are truly iron deficient,” says 

 an assistant professor of medicine (medical oncology and hematology) at Yale School of Medicine. Goshua led a team that explored the benefits and costs of screening adult women for iron deficiency and recently 

Their work is especially timely, given that the American Society of Hematology, an international task force of researchers and clinicians tackling blood disorders, convened a panel last year to establish the first medical practice guidelines for iron deficiency. The panel’s results are expected next year, and Goshua hopes his team’s work will help inform the experts’ decision.

“We believe this is how we start an important conversation that has been ignored for decades,” says Goshua, who is also a member of Yale Cancer Center.

 and 38% of reproductive age women are iron deficient. Worldwide, this issue may impact as many as 2 billion people.

“This is a problem that’s actually pretty easy to solve,” says Goshua. “We know exactly what iron does, and we know what deficiency looks like.”

Most of the body’s iron winds up in hemoglobin, a protein found in red blood cells that binds and transports oxygen to cells and tissues. Iron is absorbed from food, meaning that poor nutrition and issues with nutrient absorption due to disease or gastrointestinal disorders can deplete iron levels. Iron deficiency develops when red blood cells can’t make enough hemoglobin to meet the body’s demand for oxygen.

As a result, people often experience fatigue, shortness of breath, changes in mood, and restless legs. Goshua says these are “the kinds of things that have historically been normalized.” Part of the reason iron deficiency is underreported is because the symptoms aren’t always considered pathological.

Over time, iron deficiency can progress into anemia. But, by the time anemia presents, a person may have experienced “months or years of unnecessary symptomatology,” Goshua notes.

This is a problem that’s actually pretty easy to solve. We know exactly what iron does, and we know what deficiency looks like.
George Goshua, MD, MSc

In the new study, Goshua and his team focused on reproductive-age women, because they experience the highest rates of iron deficiency. Pregnant women are also prone to consequential iron losses, with the rate of deficiency increasing throughout pregnancy.

One recent study found that in developed countries, 84% of pregnant women experience iron deficiency during their third trimester, which can interfere with fetal brain development and deplete a baby’s baseline iron stores. Pregnant women with anemia face greater risk of preeclampsia, preterm birth, and post-partum hemorrhage.

The researchers created a model based on data from the Centers for Disease Control and Prevention’s National Health and Nutrition Examination Survey collected between 2003 and 2020 from 3,490 female patients aged 12 to 21. To test the efficacy of screening at different thresholds, they simulated three different lifetime scenarios: no screening, annual screening at 15 micrograms of ferritin per liter, and annual screening at 25 micrograms per liter. Their model captured iron loss through the reproductive years, at the median age of 51, and after menstruation ceased.

Annual iron screening at 25 micrograms per liter maximized lifetime benefits without amplifying risk such as unnecessary treatment and allergic reaction. This threshold is still conservative, Goshua notes, considering that some experts flag anyone with iron levels below 50 micrograms per liter as deficient.

Over-the-counter oral iron is the leading strategy for treating iron deficiency. It is cheap and effective but also causes constipation and abdominal pain in 40% of users. Intravenous iron is the other option and is often tolerated better. A single infusion can boost iron levels for anywhere from 10 months to multiple years, says Goshua, making it a choice treatment for people with more significant depletion or underlying health issues.

“There is a subset of folks who actually report an improvement in their symptoms within hours or days of the infusion finishing, which is pretty remarkable,” he adds.

Ultimately, the study’s findings show regular screening has clear value.

“The population health benefits are there,” says Goshua.

The Goshua Lab is supported by the National Institutes of Health (awards 1K01HL175220 and CA016359) and Yale University. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The lab is also supported by the NOMIS Foundation, the Frederick A. DeLuca Foundation, Yale Cancer Center, and Yale Bunker Endowment.

Iron deficiency affects a significant portion of the global population, particularly women of reproductive age, and can lead to anemia if untreated. Despite its prevalence, iron deficiency is often underdiagnosed due to non-standardized screening practices. A study led by Dr. George Goshua suggests that raising the ferritin threshold for screening could improve diagnosis and treatment. The study emphasizes the importance of regular screening, especially for women, and highlights the effectiveness of oral and intravenous iron treatments. The findings aim to inform upcoming guidelines by the American Society of Hematology.

Iron deficiency affects millions globally, especially women, yet remains underdiagnosed. Regular screening can prevent severe health issues and improve lives.