Predicting spatially resolved gene expression via tissue morphology using adaptive spatial GNNs.

Journal: Bioinformatics (Oxford, England)
Published Date: Sep 1, 2024

Abstract

MOTIVATION: Spatial transcriptomics technologies, which generate a spatial map of gene activity, can deepen the understanding of tissue architecture and its molecular underpinnings in health and disease. However, the high cost makes these technologies difficult to use in practice. Histological images co-registered with targeted tissues are more affordable and routinely generated in many research and clinical studies. Hence, predicting spatial gene expression from the morphological clues embedded in tissue histological images provides a scalable alternative approach to decoding tissue complexity.

Authors

  • Tianci Song
    Dept of Computer Science and Engineering, University of Minnesota Minneapolis, MN, USA.
  • Eric Cosatto
    Department of Machine Learning, NEC Laboratories America, NJ, USA.
  • Gaoyuan Wang
    Program in Computational Biology and Bioinformatics, Yale University, New Haven, CT 06520, United States.
  • Rui Kuang
    1Department of Computer Science and Engineering, University of Minnesota Twin Cities, Minneapolis, MN USA.
  • Mark Gerstein
    Program of Computational Biology and Bioinformatics and Department of Molecular Biophysics and Biochemistry and Department of Computer Science, Yale University, New Haven, CT 06511, USA.
  • Martin Renqiang Min
    Department of Machine Learning, NEC Laboratories America, Princeton, NJ 08540, USA.
  • Jonathan Warrell
    Machine Learning Department, NEC Laboratories America, Princeton, NJ 08540, United States.

Keywords

External Resources

Popular Topics
Recent Journals