How to use the application?

Input files

1. scRNA-seq data of initial cellular state. .
   
   Upload a tab-separated file with the scRNA-seq data of initial cellular state in rds/txt format. Each row represents gene and each column represents cell, which can be data frame or Seurat object (See example in Download/H9_p_E-MTAB-6819.rds).
2. Query TFs.
   
   Upload desired TFs in a tab-separated format with two columns. The first column is TF name, and the second column is the Boolean value of TFs with 1/-1, which means activation/inhibition (See example in Download /hESCp2n_TFs.txt).

Parameters

There are three parameters to be selected by the user:

• Species: human/mouse/rat.
• Perturbagen number: user can selected from 1-20. Default value is 10.
• Perturbagens category: User can specify the category of predicted perturbagens, including reprogramming, cell cycle, metabolism, immunology and FDA approved drugs. User can select more than one category. Default value is using the collected perturbagens including all categories.

Outputs

• Contextualized_PKN.rds: This file contains the contextualized PKN based on initial cellular state, mainly used for network visualization.
• Predicted_signalling_protein.xlsx: This file contains the list of predicted signalling proteins (column "Protein") and their corresponding rank (column "Rank"), sign (column "Sign") and JSD score (column "JSD_val"). The column "Sign" including values "1", "-1" and "2", which mean the corresponding protein should be activated, inhibited or unknown, respectively.
• Annotated_perturbagens.xlsx: This file contains the list of predicted perturbagens (column “Perturbagen”) and their corresponding information. Column "Target" presents the targets predicted by SiPer. Columns “Chemical_Effect” and “Predicted_Effect” shows the interaction effect between perturbagens and signalling proteins that are reported in public databases and our prediction, respectively. Value 1, -1 and 2 means activation, inhibition and unknown, respectively. Column "Group" seperates the perturbagens into different groups based on their protein targets ("Other" merges the groups with perturbagen less than 3). Users can select certain number of perturbagens (e.g., select 1or 2 perturbagens that focus on reprogramming) in each group for further experimental validation. Column "Category" presents the usability of the perturbagen in different research areas. Column "Mechanism" reports the functional annotation.