Recipes and ingredients for deep learning models of 3D genome folding.

Journal: Current opinion in genetics & development
PMID: 39862604

Abstract

Three-dimensional genome folding plays roles in gene regulation and disease. In this review, we compare and contrast recent deep learning models for predicting genome contact maps. We survey preprocessing, architecture, training, evaluation, and interpretation methods, highlighting the capabilities and limitations of different models. In each area, we highlight challenges, opportunities, and potential future directions for genome-folding models.

Authors

  • Paulina N Smaruj
    Department of Quantitative and Computational Biology, University of Southern California, Los Angeles, CA, USA.
  • Yao Xiao
    School of Energy and Environmental Engineering, Hebei University of Technology, Tianjin 300401, China; Key Laboratory of Pollution Processes and Environmental Criteria (Ministry of Education), Tianjin Engineering Center of Environmental Diagnosis and Contamination Remediation, College of Environmental Science and Engineering, Nankai University, Tianjin 300350, China.
  • Geoffrey Fudenberg
    Department of Quantitative and Computational Biology, University of Southern California, Los Angeles, CA, USA. Electronic address: fudenber@usc.edu.

Keywords

External Resources

Popular Topics
Recent Journals