Accurate single-molecule spot detection for image-based spatial transcriptomics with weakly supervised deep learning.

Journal: Cell systems
PMID:

Abstract

Image-based spatial transcriptomics methods enable transcriptome-scale gene expression measurements with spatial information but require complex, manually tuned analysis pipelines. We present Polaris, an analysis pipeline for image-based spatial transcriptomics that combines deep-learning models for cell segmentation and spot detection with a probabilistic gene decoder to quantify single-cell gene expression accurately. Polaris offers a unifying, turnkey solution for analyzing spatial transcriptomics data from multiplexed error-robust FISH (MERFISH), sequential fluorescence in situ hybridization (seqFISH), or in situ RNA sequencing (ISS) experiments. Polaris is available through the DeepCell software library (https://github.com/vanvalenlab/deepcell-spots) and https://www.deepcell.org.

Authors

  • Emily Laubscher
    Department of Chemistry, California Institute of Technology, Pasadena, CA, USA.
  • Xuefei Wang
    Department of Breast Surgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
  • Nitzan Razin
    Division of Biology and Biological Engineering, Caltech, Pasadena, CA 91125, USA.
  • Tom Dougherty
    Division of Biology and Biological Engineering, Caltech, Pasadena, CA 91125, USA.
  • Rosalind J Xu
    Program in Cellular and Molecular Medicine, Boston Children's Hospital, Boston, MA 02115, USA; Department of Microbiology, Blavatnik Institute, Harvard Medical School, Boston, MA 02115, USA; Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA 02115, USA.
  • Lincoln Ombelets
    Division of Chemistry and Chemical Engineering, Caltech, Pasadena, CA 91125, USA.
  • Edward Pao
    2Department of Bioengineering, University of California, 420 Westwood Plaza, 5121 Engineering V, PO Box 951600, Los Angeles, CA 90095 USA.
  • William Graf
    Division of Biology and Bioengineering, California Institute of Technology, Pasadena, CA, USA.
  • Jeffrey R Moffitt
    Program in Cellular and Molecular Medicine, Boston Children's Hospital, Boston, MA 02115, USA; Department of Microbiology, Blavatnik Institute, Harvard Medical School, Boston, MA 02115, USA; Broad Institute of Harvard and MIT, Cambridge, MA, USA.
  • Yisong Yue
    Department of Computing and Mathematical Sciences, California Institute of Technology, Pasadena, CA 91125, USA.
  • David Van Valen
    Division of Biology and Bioengineering, California Institute of Technology, Pasadena, CA, USA. vanvalen@caltech.edu.