image: Published today in Science Advances, co-first authors Snigdha Maiti, PhD (left), David Baggett, PhD (center left), and Swarnendu Tripathi, PhD (right), and corresponding author Richard Kriwacki, PhD (center right), St. Jude Department of Structural Biology, introduced IDR-Puncta ML, a tool that predicts which intrinsically disordered regions drive condensate formation with implications for cancer and RNA biology.
Credit: St. Jude Children's Research Hospital
(MEMPHIS, Tenn. – December 3, 2025) Fusion oncoproteins arise when a gene fuses with another gene and acquires new abilities. Such abilities can include the formation of biomolecular condensates, “droplets” of concentrated proteins, DNA or RNA. The abnormal molecular condensates formed by fusion oncoproteins can disrupt cellular functions and drive cancer development, but the specific protein features behind this process remain unclear. Scientists at St. Jude Children’s Research Hospital studied intrinsically disordered regions, unstructured protein segments that are often involved in condensate formation, to determine if they drive fusion oncoproteins to form condensates. They trained a machine learning model, called IDR-Puncta ML, with experimental data from intrinsically disordered regions in fusion oncoproteins to predict the behavior of other such regions. The model found that only about 12% of all human intrinsically disordered regions form condensates and are within proteins with strong links to RNA–related functions. Published today in Science Advances, the work provides a valuable resource for studying condensate formation in cancer and RNA biology.
Rogue biomolecular condensates are associated with multiple diseases, including neurodegeneration and aggressive pediatric cancers. Following up on a 2023 study in Nature Communications that predicted condensate formation by fusion oncoproteins, this latest work focuses on their intrinsically disordered regions, which had been generally linked to condensate formation. The findings reveal that biomolecular condensate formation exclusively through intrinsically disordered regions occurs only in a small group of specialized proteins, underscoring the complexity of this process across different molecular contexts.
“We reasoned from previous experiments that intrinsically disordered regions were associated with condensate formation by a significant portion of droplet-forming fusion oncoproteins,” said corresponding author of the study Richard Kriwacki, PhD, St. Jude Department of Structural Biology. “This allowed us to apply our data science tools to understand the sequence features underlying these results, both within fusion oncoproteins and human proteins in general, providing insight into the role of these flexible protein regions in human biology.”
Small percentage of intrinsically disordered regions form condensates
The team combined machine learning with a robust experimental pipeline to develop IDR-Puncta ML. “We built a dataset by testing different intrinsically disordered regions from various fusions and experimentally validated whether they could independently form droplets, or puncta, in cells,” said co-first author Snigdha Maiti, PhD, Department of Structural Biology. “Based on this dataset, we created a machine learning model to predict if other intrinsically disordered regions with similar amino acid sequence features could also form condensates in cells.”
Their model showed high accuracy — over 90% — in condensate prediction when independent testing was performed, validating their findings and allowing them to extend their predictions to all known intrinsically disordered regions in the human proteome. Of note, just 12% were predicted to form condensates, a finding supported by similar studies in other labs.
“This frequency is low when you consider over 60% of fusion oncoproteins likely form condensates,” said co-first author Swarnendu Tripathi, PhD, Department of Structural Biology. “This suggests that condensate-forming intrinsically disordered regions are likely related to specific functions.”
This hypothesis was confirmed when the team dug into the positive predictions and found that those proteins were largely tied to RNA biology. “We found that intrinsically disordered regions driving condensate formation are within proteins that have specific cellular functions, such as RNA processing, splicing and regulation of RNA metabolism, but are rarer than expected,” said co-first author David Baggett, PhD, Department of Structural Biology. “This implies that while some intrinsically disordered regions can form condensates independently, others might need help from other regions of the protein to form condensates.”
IDR-Puncta ML is freely available to use at https://sak.stjude.org/ and provides a vital platform to better understand the molecular mechanisms behind condensate formation and its links to disease. “Understanding how fusion oncoproteins drive cancer is the first step toward developing treatments, because we can’t treat what we don’t understand,” Baggett said. “We’re getting to the core of why these proteins, and the condensates they form, have the effect they do, and that’s the first step to correcting it.”
Authors and funding
The study’s other authors are Swati Kinger and Ramiz Somjee, Rhodes College and St. Jude; and Aaron Philips, Cheon-Gil Park, Jina Wang, Wahiduzzaman, William Freyaldenhoven, Brittany Pioso, John Bollinger, Benjamin Lang, and M. Madan Babu, St. Jude.
The study was supported by the National Cancer Institute (R01 CA246125, U54 CA243124 and P30 CA021765) and the American Lebanese Syrian Associated Charities (ALSAC), the fundraising and awareness organization of St. Jude.
St. Jude Media Relations Contact
Chelsea Bryant
Desk: (901) 595-0564
Cell: (256) 244-2048
chelsea.bryant@stjude.org
media@stjude.org
St. Jude Children’s Research Hospital
St. Jude Children’s Research Hospital is leading the way the world understands, treats, and cures childhood catastrophic diseases. From cancer to life-threatening blood disorders, neurological conditions, and infectious diseases, St. Jude is dedicated to advancing cures and means of prevention through groundbreaking research and compassionate care. Through global collaborations and innovative science, St. Jude is working to ensure that every child, everywhere, has the best chance at a healthy future. To learn more, visit stjude.org, read St. Jude Progress, a digital magazine, and follow St. Jude on social media at @stjuderesearch.
Journal
Science Advances
Article Title
Proteome-wide computational analyses reveal links between protein condensate formation and RNA biology
Article Publication Date
3-Dec-2025