AI-Powered Genome Mapping Reveals New Layer of Alzheimer’s Disease Biology

Clarifying the molecular architecture underlying neuronal dysfunction remains a major challenge, limiting progress on precise biomarkers and targeted interventions. While amyloid-beta plaques and tau tangles are classic features, additional regulatory layers likely contribute to disease biology. A new study shows that single-cell three-dimensional (3D) genome mapping, combined with spatial transcriptomics and artificial intelligence, links chromatin organization to gene activity and tissue pathology in Alzheimer’s disease.

Researchers at the University of Pittsburgh School of Medicine and Carnegie Mellon University detailed a multimodal approach that integrates single-cell 3D genome mapping, spatial transcriptomics, and deep learning. The work identifies “increased compartment mingling,” in which large active and inactive genomic regions are less clearly separated in Alzheimer’s brain cells than in cells without disease. The team also introduced Hicformer, an artificial intelligence model designed to predict gene activity by incorporating DNA sequence, broad genome-folding features, and local 3D contact maps. Together, these tools enabled a multiscale view of genome architecture, transcriptional programs, and brain tissue context.

The methodology centers on GAGE-seq, which simultaneously measures gene expression and 3D genome contacts within the same cell. Researchers integrated these single-cell measurements with spatial transcriptomic maps of intact brain tissue to relate higher-order chromatin organization to local gene regulation. This framework allowed disease-associated molecular and cellular features to be placed within their anatomical tissue environment.

Analyses revealed fewer short-range and more long-range genomic contacts across multiple brain cell types in Alzheimer’s disease, with greater compartment mingling associated with lower overall gene activity. Contacts between genes and regulatory elements weakened, while some midrange contacts strengthened. These architectural alterations were linked to reduced neuronal and synaptic programs, altered metabolic and stress responses, and stabilization in microglia. Published in Science, the findings establish higher-order chromatin alterations as part of Alzheimer’s molecular pathology and connect genome folding to tissue-level changes.

Postmortem prefrontal cortex tissue was obtained from individuals with and without Alzheimer’s disease who had participated in a long-term study at Rush University’s Alzheimer’s Disease Center and donated their brains after death. Collaborating institutions included the Broad Institute of MIT and Harvard, the University of California, Los Angeles, and the University of Washington. According to the team, the technology could be applied to numerous diseases; the group focused on Alzheimer’s because of its rising prevalence, and the discovery raises new questions about causality, susceptibility, asymptomatic pathology, and potential therapeutic targets.

“Alzheimer’s disease cannot be understood one layer at a time. The genome’s 3D structure is a fundamental regulatory layer that helps to connect DNA sequence to gene activity. By integrating genome folding, cell state and tissue context, we can move beyond cataloging disease-associated changes toward understanding how they fit together and which mechanisms to test next,” said Jian Ma, the Ray and Stephanie Lane Professor of Computational Biology at Carnegie Mellon University, who led and supervised the study.

“We know the classic hallmarks of Alzheimer’s disease—accumulation of amyloid-beta plaques and tau tangles—but our results establish higher-order chromatin alterations as a component of the molecular pathology associated with the disease, which currently affects seven million Americans, a number that continues to grow,” said Hansruedi Mathys, assistant professor of neurobiology at the University of Pittsburgh and cosenior author of the Science paper.

Related Links
University of Pittsburgh School of Medicine
Carnegie Mellon University