EUVIMEDEUVIMED entdecken
Open Access

Genome-wide methylome modeling via generative AI incorporating long- and short-range interactions

Fengyao Yan iD, Aristeidis G Telonis iD, Qin Yang iD, Limin Jiang, Francine E Garrett-Bakelman iD, Mikkael A Sekeres iD, Valeria Santini iD, Michele Ceccarelli iD, Neha Goel iD, Liliana Garcia-Martinez iD, Lluis Morey iD, Maria E Figueroa iD und 1 weitere

Science Advances · 2025

DOI: 10.1126/sciadv.adt4152 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Advances, 2025. Verfasst von Fengyao Yan und 12 weiteren Autor:innen. Beteiligte Einrichtung: Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine, Miami, FL 33136, USA. Hauptthemen laut MEDLINE: DNA Methylation, Epigenome, Artificial Intelligence, Genome, Human. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS, NCI NIH HHS, NIEHS NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Using millions of methylation segments, we developed DiffuCpG, a generative artificial intelligence (AI) diffusion model designed to solve the critical challenge of missing data in high-throughput methylation technologies. DiffuCpG goes beyond conventional methods by leveraging both short-range interactions including nearby CpGs from both latitude and longitude of the dataset, local DNA sequences, and long-range interactions, including three-dimensional genome architecture and long-distance correlations, to comprehensively model the methylome. Compared to previous methods, through extensive independent validations across different tissue types, cancers, and technologies (whole-genome bisulfite sequencing, enhanced reduced representation bisulfite sequencing, single-cell bisulfite sequencing, and methylation arrays), DiffuCpG has demonstrated superior performance in accuracy, scalability, and versatility. On average, bisulfite sequencing dataset, DiffuCpG can extend the original dataset by millions of additional CpGs. As an alternative application of generative AI, DiffuCpG addresses a key bottleneck in epigenetic research and will substantially benefit studies relying on high-throughput methylation data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

DNA MethylationEpigenomeArtificial IntelligenceGenome, HumanEpigenomicsHumansCpG IslandsEpigenesis, GeneticHigh-Throughput Nucleotide SequencingAlgorithms

Autor:innen und Institutionen

  • Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine, Miami, FL 33136, USA.
  • Department of Biochemistry and Molecular Biology, University of Miami Miller School of Medicine, Miami, FL 33136, USA.
  • Department of Human Genetics, University of Miami Miller School of Medicine, Miami, FL 33136, USA.
  • Department of Biochemistry and Molecular Genetics, University of Virginia, Charlottesville, VA 22908, USA.
  • Department of Medicine, University of Virginia, Charlottesville, VA 22908, USA.
  • Comprehensive Cancer Center, University of Virginia, Charlottesville, VA 22908, USA.

ORCID-Profile: Fengyao Yan, Aristeidis G Telonis, Qin Yang, Francine E Garrett-Bakelman, Mikkael A Sekeres, Valeria Santini, Michele Ceccarelli, Neha Goel, Liliana Garcia-Martinez, Lluis Morey, Maria E Figueroa, Yan Guo

Förderung

  • NIGMS NIH HHS
  • NCI NIH HHS
  • NIEHS NIH HHS

Zitieren

APA

Yan, F., G Telonis, A., Yang, Q., Jiang, L., E Garrett-Bakelman, F., Sekeres, M. A., Santini, V., Ceccarelli, M., Goel, N., Garcia-Martinez, L., Morey, L., Figueroa, M. E., & Guo, Y. (2025). Genome-wide methylome modeling via generative AI incorporating long- and short-range interactions. Science Advances. https://doi.org/10.1126/sciadv.adt4152

Vancouver

Yan F, G Telonis A, Yang Q, Jiang L, E Garrett-Bakelman F, Sekeres MA, et al. Genome-wide methylome modeling via generative AI incorporating long- and short-range interactions. Science Advances. 2025. doi:10.1126/sciadv.adt4152

BibTeX
@article{Yan2025,
  author = {Yan, Fengyao and G Telonis, Aristeidis and Yang, Qin and Jiang, Limin and E Garrett-Bakelman, Francine and Sekeres, Mikkael A and Santini, Valeria and Ceccarelli, Michele and Goel, Neha and Garcia-Martinez, Liliana and Morey, Lluis and Figueroa, Maria E and Guo, Yan},
  title = {Genome-wide methylome modeling via generative AI incorporating long- and short-range interactions},
  journal = {Science Advances},
  year = {2025},
  doi = {10.1126/sciadv.adt4152}
}
Kostenlos starten →