SPADE

Integrating spatial transcriptomic data with deep features of tissue image

Requirements

python >=3.7
tensorflow >= 1.14.0
keras >= 2.3

R >=3.6.1
Seurat >=3.1.2. (For Spatial Transcriptome Data)
limma >=3.42.

How to Run

Command Line

python spade_spatial_marker_by_deeplearning.py --position [Tissue Position List File] --image [High Res Image File] --scale [Scale for High Res Image] --meta [metadata csv file] --outdir [Output directory]

• Files Info
  Tissue Position List file : Tissue coordinate file includes barcodes, row and col coordinates.
  High Res Image File : PNG, JPEG or TIFF for high resolution tissue images
  Scale : Scale for high resolution image. For Visium, find from a scalefactos_json file.
  meta : meta data for barcodes and clusters. \

• Example
  Check breastca_spatial_SPADE.R file for an example.\

Reference

Sungwoo Bae, Hongyoon Choi, Dong Soo Lee, bioRxiv 2020.06.15.150698; doi: https://doi.org/10.1101/2020.06.15.150698