AI Revealed How DNA Architecture Failures Can Trigger Blood Cancer, Researchers Report at ASH 2025
- camilarosiaz
- 5 days ago
- 2 min read
Artificial intelligence helped researchers uncover how subtle failures in the three-dimensional structure of DNA can set the stage for blood cancer, according to findings presented on December 6, 2025, at the American Society of Hematology annual meeting.
Image source Depositphotos.
The study was led by Martin Rivas, Ph.D., a cancer researcher at Sylvester Comprehensive Cancer Center, part of the University of Miami Miller School of Medicine. His team showed that disruptions in genome architecture—rather than direct genetic mutations alone—can predispose cells to lymphoma.
The research introduced the concept of architectural tumor suppression, focusing on proteins such as SMC3 and CTCF, which help organize DNA into loops that connect regulatory elements known as enhancers with the genes they activate. When only one functional copy of these proteins is present, a condition known as haploinsufficiency, critical DNA loops are lost, leading to the silencing of tumor suppressor genes.
Using AI-driven analysis, the team integrated large-scale datasets from Hi-C genome mapping, single-cell RNA sequencing, and epigenetic profiling. The approach revealed that partial loss of SMC3 or CTCF did not collapse the entire genome structure but selectively disrupted short-range enhancer–promoter loops. These changes affected tumor suppressor genes such as Tet2, Kmt2d, and Dusp4.
The architectural breakdown also interfered with normal B-cell development, preventing cells from maturing into plasma cells and creating conditions that favor malignancy. Clinical data supported the findings, showing that patients with diffuse large B-cell lymphoma who had lower levels of SMC3 expression experienced poorer outcomes.
Dr. Martin Rivas, a cancer researcher at Sylvester Comprehensive Cancer Center.
The results suggested that genome architecture could serve as both a prognostic biomarker and a future therapeutic target. Rather than correcting mutations alone, emerging strategies could focus on restoring or mimicking lost DNA loops to maintain proper gene regulation.
More research updates from Sylvester are available on the InventUM blog, with ongoing findings shared on X via @SylvesterCancer.