University of Maryland Professor Pratyush Tiwary is part of a consortium awarded up to $51.7 million by the Advanced Research Projects Agency for Health (ARPA-H) to predict how disease-causing proteins misfold before irreversible damage occurs, and to use those predictions to develop new diagnostics and treatments. The four-year effort, announced October 12, 2026, will initially focus on Alzheimer’s disease and Ewing sarcoma.
Called SEEDS (Sequence-to-disease: Environment-aware Ensembles of Disordered proteins at Scale), the consortium is co-led by Priyanka Joshi of Georgetown University and Chuankai (Kai) Zhou of the Buck Institute for Research on Aging. It aims to generate the data needed to train artificial intelligence to recognize the rules governing disease-causing proteins and predict their behavior. Many of these proteins have long been considered “undruggable” because they lack the stable shapes and binding pockets most drugs are designed to fit.
Tiwary’s Role: Chemistry-First AI
Tiwary, who holds appointments at the University of Maryland, College Park and the University of Maryland, Baltimore, will serve as a co-principal investigator on INTREPID (INTegRated Environment-aware Prediction Engine for Intrinsically Disordered Protein Pathology), the Georgetown-led part of the consortium. INTREPID will examine how a protein’s sequence and cellular environment shape the range of forms it takes and how it behaves, combining high-throughput and high-resolution experimental measurements with computation and AI.
“At the heart of INTREPID is a chemistry-first generative AI model that predicts how disordered protein ensembles change across cellular conditions such as pH, crowding, metabolites, oxidative stress and post-translational modifications,” Tiwary said.
The team will start with tau and EWS::FLI1, intrinsically disordered proteins implicated in Alzheimer’s disease and Ewing sarcoma, respectively. Unlike proteins with relatively stable 3D structures, intrinsically disordered proteins take many shapes depending on cellular conditions. That flexibility serves important functions in healthy cells but makes the proteins extraordinarily difficult to understand and target when something goes wrong.
“What makes this award especially exciting is that the AI model will be experimentally validated, guide drug discovery, and ultimately help power an inexpensive, accurate approach for the diagnosis and treatment of deadly diseases,” said Tiwary, the Millard and Lee Alexander Professor in Chemical Physics in UMD’s Department of Chemistry and Biochemistry and Institute for Physical Science and Technology, and director of therapeutic drug discovery at the University of Maryland Institute for Health Computing.
Filling AI’s Data Gap
AI tools such as AlphaFold have transformed protein science by learning from decades of structural data, but researchers lack a comparable body of data for intrinsically disordered proteins, Zhou said, making them much harder to model. SEEDS pairs computational predictions with experimental measurements, testing protein behavior under different conditions rather than relying on computer models alone.
“This work supports my lab’s efforts to thoughtfully and effectively develop new AI approaches specifically for chemistry and not just apply existing AI methods to these problems,” said Tiwary, who recently led a National Science Foundation-funded workshop on building chemistry-first AI tools designed around the field’s distinctive data, questions, and experimental practices.
Learning the “Grammar” of Protein Behavior
With researchers from 10 academic institutions and seven industry partners, SEEDS is part of ARPA-H’s BIOmolecular Grammar for protein Aggregation Modulation and Intervention (BIOGAMI) program, led by Shannon Greene. BIOGAMI focuses on the earliest stages of protein dysfunction, before proteins misfold or clump together and cause lasting damage, using the idea of a molecular “grammar” whose rules could let researchers predict and influence when protein behavior begins to change.
Alongside INTREPID, a second effort called SHIFT (Structural Characterization of Intrinsically Disordered Regions at Scale) will map the context-dependent shapes intrinsically disordered regions take across a broad range of human proteins. Data from both will train foundation models designed to predict those protein ensembles.
“We aim to uncover the principles that determine when these proteins remain in healthy functional ensembles and when they begin to shift toward disease-associated states,” Zhou said. “Ultimately, this work could enable us to detect and potentially redirect harmful conformational changes before they lead to irreversible cellular damage.”
“We want to catch the seed before it grows into disease,” Joshi said.
About the University of Maryland College of Computer, Mathematical, and Natural Sciences
The College of Computer, Mathematical, and Natural Sciences at the University of Maryland educates more than 10,000 future scientific leaders in its undergraduate and graduate programs each year. The college’s 10 departments and seven interdisciplinary research centers foster scientific discovery, with annual sponsored research funding exceeding $250 million.
About ARPA-H
The Advanced Research Projects Agency for Health (ARPA-H), an agency within the U.S. Department of Health and Human Services, funds bold, high-impact research to solve some of the most challenging problems in health. This research was funded in part by ARPA-H. The views and conclusions it contains are those of the authors and should not be interpreted as representing the official policies, expressed or implied, of the U.S. Government.
