ORION Whole Slide Multiplexed Imaging

ORION is a method for collecting one-shot and multi-cycle 18-plex immunofluorescence images and diagnostic-grade H&E images from the same samples. The Orion method was developed in collaboration with RareCyte Inc. and uses a specialized microscope and fluorescent antibodies, which can be imaged simultaneously and spectrally unmixed. Same-slide H&E and IF images provide complementary information that can be used to train ML models to predict cancer progression.

For the Breast Atlas, ORION is performed to visualize distinct markers and quantify the expression and spatial heterogeneity of markers across ROIs corresponding to normal breast and benign regions.

About the Data Data Levels

About the Data


Method and Protocol

Learn about the Orion method by reading:

Data Visualization

Full-resolution images can be viewed in a web browser using Minerva:

  • via the gallery (forthcoming)
  • in cBioPortal (forthcoming)

Learn more about the Minerva software at minerva.im.

Data Access

All image data will be accessible from Amazon Web Services (AWS):

  • Levels 1-2: OME-TIFF
  • Level 3: OME-TIFF
  • Level 4: csv

For a description of the files see the table below.

About the Samples

The following are planned to image with Orion:

  • 50-100 normal breast samples
  • 75 ovarian samples

Instruments

Orion data was collected on RareCyte Orion 6 and 11 microscopes.

About the Data Generators

This data was generated and analyzed by the Laboratory of Systems Pharmacology at Harvard Medical School.


Data Levels:

Orion uses OME-TIFF and other BioFormats file formats and follows MITI standards.

Data Type Description File Format Average size (per sample) Data Location
High-plex image data (Levels 1-2) Quality controlled images are assembled into a high-plex whole slide image OME-TIFF 100 GB AWS
Segmentation masks (Level 3) Cell segmentation identifies and separates cell, nucleus, cytoplasm and other regions OME-TIFF 0.5 GB AWS
Single-cell spatial feature tables (Level 4) Summarize the expression of marker intensities and morphological features for each cell csv 1 GB AWS