New AI Tool LogiRx Accelerates Drug Repurposing for Disease Treatment

New artificial intelligence tool accelerates disease treatments

University of Virginia School of Medicine scientists have created a computational tool to accelerate the development of new disease treatments. The tool goes beyond current artificial intelligence (AI) approaches by identifying not just which patient populations may benefit but also how the drugs work inside cells. The researchers have demonstrated the tool's potential by identifying a promising candidate to prevent heart failure, a leading cause of death in the United States and around the world. The new AI tool called LogiRx, can predict how drugs will affect biological processes in the body, helping scientists understand the effects the drugs will have other than their original purpose. For example, the researchers found that the antidepressant escitalopram, sold as Lexapro, may prevent harmful changes in the heart that lead to heart failure, a condition that causes almost half of all cardiovascular deaths in the United States. "AI needs to move from detecting patterns to generating understanding," said UVA's Jeffrey J. Saucerman. PhD. "Our LogiRx tool helps us identify not just which drugs can be repurposed for heart disease but how they work in the heart." Preventing Heart Failure Heart failure kills more than 400,000 Americans every year. One of its hallmarks is the overgrowth of cells that thicken the heart muscle and prevent the organ from pumping blood as it should. This is known as cardiac hypertrophy. Saucerman and his team, led by PhD student Taylor Eggertsen, wanted to see if LogiRx could identify drugs with the potential to prevent cardiac hypertrophy and, ultimately, head off heart failure. They used the tool to evaluate 62 drugs that had been previously identified as promising candidates for the task. LogiRx was able to predict "off-target" effects for seven of these drugs that could help prevent harmful cellular hypertrophy, which were confirmed in cells for two of the drugs. The scientists then evaluated LogiRx's predictions by doing lab tests and by looking at outcomes in patients taking the drugs. The latter revealed that patients taking escitalopram were significantly less likely to develop cardiac hypertrophy. "LogiRx identifies unexpected new uses for old drugs that are already shown to be safe in humans," said Eggertsen, in UVA's Department of Biomedical Engineering, a joint program of the School of Medicine and School of Engineering. "This tool can help researchers explore new patient populations that could benefit from a drug or to avoid unwanted side effects." Additional lab research and clinical trials will be needed before doctors might start prescribing escitalopram for heart health. But Saucerman is excited about the potential of LogiRx for advancing and accelerating new treatments not just for cardiac hypertrophy but for a host of other serious medical conditions. "AI is accelerating many aspects of drug development, but it has made less progress in providing the required understanding of how these drug work in the body," Saucerman said. "LogiRx is a step towards combining AI with existing knowledge of how cells work to find new uses for old drugs."

UVA scientists develop AI tool to accelerate the development of new disease treatments

University of Virginia Health SystemMar 18 2025 University of Virginia School of Medicine scientists have created a computational tool to accelerate the development of new disease treatments. The tool goes beyond current artificial intelligence (AI) approaches by identifying not just which patient populations may benefit but also how the drugs work inside cells. The researchers have demonstrated the tool's potential by identifying a promising candidate to prevent heart failure, a leading cause of death in the United States and around the world. The new AI tool called LogiRx, can predict how drugs will affect biological processes in the body, helping scientists understand the effects the drugs will have other than their original purpose. For example, the researchers found that the antidepressant escitalopram, sold as Lexapro, may prevent harmful changes in the heart that lead to heart failure, a condition that causes almost half of all cardiovascular deaths in the United States. AI needs to move from detecting patterns to generating understanding. Our LogiRx tool helps us identify not just which drugs can be repurposed for heart disease but how they work in the heart." Jeffrey J. Saucerman. PhD, UVA Preventing heart failure Heart failure kills more than 400,000 Americans every year. One of its hallmarks is the overgrowth of cells that thicken the heart muscle and prevent the organ from pumping blood as it should. This is known as cardiac hypertrophy. Saucerman and his team, led by PhD student Taylor Eggertsen, wanted to see if LogiRx could identify drugs with the potential to prevent cardiac hypertrophy and, ultimately, head off heart failure. They used the tool to evaluate 62 drugs that had been previously identified as promising candidates for the task. LogiRx was able to predict "off-target" effects for seven of these drugs that could help prevent harmful cellular hypertrophy, which were confirmed in cells for two of the drugs. The scientists then evaluated LogiRx's predictions by doing lab tests and by looking at outcomes in patients taking the drugs. The latter revealed that patients taking escitalopram were significantly less likely to develop cardiac hypertrophy. "LogiRx identifies unexpected new uses for old drugs that are already shown to be safe in humans," said Eggertsen, in UVA's Department of Biomedical Engineering, a joint program of the School of Medicine and School of Engineering. "This tool can help researchers explore new patient populations that could benefit from a drug or to avoid unwanted side effects." Additional lab research and clinical trials will be needed before doctors might start prescribing escitalopram for heart health. But Saucerman is excited about the potential of LogiRx for advancing and accelerating new treatments not just for cardiac hypertrophy but for a host of other serious medical conditions. "AI is accelerating many aspects of drug development, but it has made less progress in providing the required understanding of how these drug work in the body," Saucerman said. "LogiRx is a step towards combining AI with existing knowledge of how cells work to find new uses for old drugs." Findings published The researchers have published their findings in PNAS, the Proceedings of the National Academy of Sciences. The research team consisted of Eggertsen, Joshua G. Travers, Elizabeth J. Hardy, Matthew J. Wolf, Timothy A. McKinsey and Saucerman. The scientists have no financial interest in the work. Source: University of Virginia Health System Journal reference: Eggertsen, T. G., et al. (2025). Logic-based machine learning predicts how escitalopram attenuates cardiomyocyte hypertrophy. Proceedings of the National Academy of Sciences. doi.org/10.1073/pnas.2420499122.

AI tool predicts how body will respond to drugs, potentially accelerating disease treatments

University of Virginia School of Medicine scientists have created a computational tool to accelerate the development of new disease treatments. The tool goes beyond current artificial intelligence (AI) approaches by identifying not just which patient populations may benefit but also how the drugs work inside cells. The researchers have demonstrated the tool's potential by identifying a promising candidate to prevent heart failure, a leading cause of death in the United States and around the world. The research is published in the journal Proceedings of the National Academy of Sciences. The new AI tool called LogiRx, can predict how drugs will affect biological processes in the body, helping scientists understand the effects the drugs will have other than their original purpose. For example, the researchers found that the antidepressant escitalopram, sold as Lexapro, may prevent harmful changes in the heart that lead to heart failure, a condition that causes almost half of all cardiovascular deaths in the United States. "AI needs to move from detecting patterns to generating understanding," said UVA's Jeffrey J. Saucerman. Ph.D. "Our LogiRx tool helps us identify not just which drugs can be repurposed for heart disease but how they work in the heart." Preventing heart failure Heart failure kills more than 400,000 Americans every year. One of its hallmarks is the overgrowth of cells that thicken the heart muscle and prevent the organ from pumping blood as it should. This is known as cardiac hypertrophy. Saucerman and his team, led by Ph.D. student Taylor Eggertsen, wanted to see if LogiRx could identify drugs with the potential to prevent cardiac hypertrophy and, ultimately, head off heart failure. They used the tool to evaluate 62 drugs that had been previously identified as promising candidates for the task. LogiRx was able to predict "off-target" effects for seven of these drugs that could help prevent harmful cellular hypertrophy, which were confirmed in cells for two of the drugs. The scientists then evaluated LogiRx's predictions by doing lab tests and by looking at outcomes in patients taking the drugs. The latter revealed that patients taking escitalopram were significantly less likely to develop cardiac hypertrophy. "LogiRx identifies unexpected new uses for old drugs that are already shown to be safe in humans," said Eggertsen, in UVA's Department of Biomedical Engineering, a joint program of the School of Medicine and School of Engineering. "This tool can help researchers explore new patient populations that could benefit from a drug or to avoid unwanted side effects." Additional lab research and clinical trials will be needed before doctors might start prescribing escitalopram for heart health. But Saucerman is excited about the potential of LogiRx for advancing and accelerating new treatments not just for cardiac hypertrophy but for a host of other serious medical conditions. "AI is accelerating many aspects of drug development, but it has made less progress in providing the required understanding of how these drug work in the body," Saucerman said. "LogiRx is a step towards combining AI with existing knowledge of how cells work to find new uses for old drugs." The research team consisted of Eggertsen, Joshua G. Travers, Elizabeth J. Hardy, Matthew J. Wolf, Timothy A. McKinsey and Saucerman. The scientists have no financial interest in the work.