SCTCwhatateam: Predicting cell locations based on location-marker genes

Supplementary folder: https://github.com/pvvhoang/SCTCwhatateam-ShinyApp/tree/master/Supplementary

1. Using a package

The Python functions (linGen, linFwd, linRev) could be retrieved by running the linmethods.py provided in the Supplementary folder.

library(reticulate)
# source_python("./linmethods.py")

The data example is provided by DREAM Single Cell Transcriptomics Challenge in the Supplementary folder.

2. Shiny app

Please see tutorial in the "Shiny app tutorial.docx" in the Supplementary folder.

To simplify we have developed an easy-to-use user interface at https://github.com/pvvhoang/SCTCwhatateam-ShinyApp

The input requirements are based on each method's preference. In general,

• Reference gene expression for locations (Locations x Genes): see bdtnp.csv

• Reference scRNA-Seq binarized file (Locations x Genes): see binarized_bdtnp.csv

• Raw scRNA-Seq expression (Cell x Genes): see dge_raw.csv

• scRNA-Seq normalized file (Cells x Genes): see dge_normalized.csv

• 3D coordinates for each locations (Locations x (x,y,z) coordinates): see geometry.csv

• Seed list (Locations x coordinates): seed_list.csv