Supercomputing powers breakthrough blood-clot research

Red blood cells may be tiny, but the physics governing their flow is anything but simple. Thanks to U.S. National Science Foundation (NSF) ACCESS allocations on the Texas Advanced Computing Center’s Stampede3 supercomputer, researchers are mapping this complexity across scales to shed new light on clot formation. Blood flow not only moves vital products through the body, it also performs healing functions like blood clotting, which when it goes wrong can lead to strokes and heart attacks, the leading cause of death in people older than 65.

Z. Leonardo Liu, an assistant professor at Florida State University and the FAMU-FSU College of Engineering, has been using ACCESS (formerly XSEDE and TeraGrid) systems since 2010 to explore fundamental questions about the flow of suspended particles like red blood cells. His recent work used ACCESS allocations on Stampede3 to investigate how fluid forces in circulating blood influence clotting at the molecular and cellular scale.

A major focus of Liu’s research is the von Willebrand factor (VWF), a large protein essential to clot formation. Studying VWF helps address two major health threats. Excessive levels or hyperactivity of VWF can cause occlusive clots, which completely block blood vessels, leading to heart attacks and strokes. Deficiencies or abnormalities in VWF are also a problem and can lead to bleeding disorders or the failure to form clots during major blood loss, a leading cause of death in people younger than 45.

"Simulating VWF interactions in crowded blood flow is a memory-bandwidth bound problem and Stampede3’s High Bandwidth Memory (HBM) was a critical fit, allowing data transfer speeds that keep up with the chaotic, rapid movement of cells and protein under high shear,” Liu said. “As an HPC user since the Ranger era, I have seen TACC consistently evolve its hardware to match the complexity of the science. Moving from simple fluids to living biological systems required the hardware evolution (from standard CPUs to HBM) that TACC has reliably provided via NSF-funded allocations like ACCESS."

Liu said that ACCESS enabled his team to simulate realistic cellular blood flow environment with full physiological fidelity, increasing their effective problem size by more than 100 times without increasing time-to-solution. 

“Replicating these computationally intensive, particle-based simulations on local desktop platforms would have cost years to obtain useful data,” Liu said.

The integration of computational modeling with experimental validation that characterizes Liu’s research methodology represents the cutting-edge interdisciplinary approach that defines modern biomedical engineering. His work demonstrates how engineering principles can provide fundamental insights into biological processes that directly impact human health outcomes. By integrating advanced supercomputing simulations with whole-blood experiments, his team investigates how proteins and cells interact in the flow of blood affected by disease or injury, an area that has traditionally been difficult to probe with conventional research methods.

The primary computational challenge is the multiscale coupling required to simulate VWF polymers unraveling within crowded, high-shear cellular flows. The Liu Lab needed to resolve hydrodynamics, cell membrane mechanics, and macromolecule conformational changes simultaneously to capture the physical events that trigger clotting. 

"ACCESS provided the massive parallelization necessary to couple millions of Lagrangian points (cells and proteins) with Eulerian fluid solvers,” Liu said. “This allowed us to resolve the fine temporal scales needed to capture split-second molecular activation events that standard computing resources cannot handle." 

Liu’s work has revealed surprising insights about the role of red blood cells in blood clotting regulation. His recent studies show that blood clotting proteins operate like tiny mechanical switches, toggling on or off in response to precisely controlled mechanical forces generated by the collective motion of red blood cells.

“Our findings show that red blood cells, once thought of merely as passive carriers of oxygen, also play an active role in regulating these ‘tiny switches’ that control blood clotting through intricate fluid-mechanical forces,” Liu said.

In 2025, Z. Leonardo Liu was awarded the 2024 Eberhard F. Mammen Young Investigator Award for his groundbreaking research into blood clotting mechanisms.

Liu is the only awardee from the U.S. among six international recipients, placing him among the most promising young researchers advancing the understanding of thrombosis and hemostasis, the complex biological processes that control blood clotting and bleeding.

“I was both surprised and deeply honored to receive this recognition,” Liu said. “Being acknowledged by such a respected journal and editorial board is incredibly meaningful. I hope this visibility will expand the reach of our research and open doors to new collaborations that lead to transformative therapies for blood-related diseases.”

"Supercomputers act as a 'computational microscope' for the nanomechanical triggers of disease. They allow us to watch how individual blood cells physically stretch and activate clotting proteins in extreme flows. By making these invisible mechanical forces visible, we can understand exactly how VWF turns from 'safe' to 'sticky,' leading to better drugs and safer medical devices."