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

Emily Laubscher; Xuefei Wang; Nitzan Razin; Tom Dougherty; Rosalind J Xu; Lincoln Ombelets; Edward Pao; William Graf; Jeffrey R Moffitt; Yisong Yue; David Van Valen

doi:10.1016/j.cels.2024.04.006

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

Cell Syst. 2024 May 15;15(5):475-482.e6. doi: 10.1016/j.cels.2024.04.006.

Authors

Affiliations

¹ Division of Chemistry and Chemical Engineering, Caltech, Pasadena, CA 91125, USA.
² Division of Biology and Biological Engineering, Caltech, Pasadena, CA 91125, USA.
³ 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.
⁴ 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.
⁵ Division of Computational and Mathematical Sciences, Caltech, Pasadena, CA 91125, USA.
⁶ Division of Biology and Biological Engineering, Caltech, Pasadena, CA 91125, USA; Howard Hughes Medical Institute, Chevy Chase, MD 20815, USA. Electronic address: [email protected].

PMID: 38754367
DOI: 10.1016/j.cels.2024.04.006

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.

Keywords: deep learning; image analysis; spatial transcriptomics; spot detection.

MeSH terms

Animals
Deep Learning*
Gene Expression Profiling* / methods
Humans
Image Processing, Computer-Assisted / methods
In Situ Hybridization, Fluorescence* / methods
Single Molecule Imaging / methods
Single-Cell Analysis / methods
Software
Supervised Machine Learning
Transcriptome* / genetics