Back

Repetitive DNA-resolved epigenomics maps chromatin rewiring during aging and malignant transformation

Alba-Linares, J. J.; Tejedor, J. R.; Sainz-Ledo, L.; Fernandez, A. F.; Perez, R. F.; Fraga, M. F.

2026-05-01 molecular biology
10.64898/2026.04.29.721644 bioRxiv
Show abstract

Repetitive DNA constitutes more than half of the human genome, yet its chromatin landscape remains poorly understood. Here, we integrated multi-mapping reads across 3, 744 repeat subfamilies and 937 ChIP-seq experiments for six canonical histone marks in hematopoietic cells and chronic malignancies from the BLUEPRINT consortium. Chromatin signals at repeats encode cell type-specific signatures, with certain elements (Alu, SVA, ERV/Gypsy LTRs) exhibiting conserved epigenomic profiles distinct from surrounding single-copy sequences, and LTRs acting as chromatin boundaries. Combinatorial analyses further revealed five major repeat epitypes (promoter-like, active/poised enhancers, and two heterochromatin domains) that enhance annotations from uniquely-mapping reads. Beyond constitutive heterochromatin, we uncovered a highly plastic heterochromatin characterized by concurrent H3K36me3 and H3K9me3 that can dynamically switch to active epitypes in B-cell chronic neoplasms. Although malignant transformation involved global loss of histone marks across repeats, we identified targeted heterochromatinization of satellites and LTRs, mirroring changes in aged B cells, and gains of active marks at CD34 enhancers and disease-related genes. Likewise, aging primes the heterochromatin landscape toward malignancy but still requires de novo acquisition of active marks at regulatory regions to promote transformation. Our atlas provides the most comprehensive resource for dissecting chromatin dynamics at repetitive DNA in hematopoietic aging and cancer.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.