The compressed file can be either tar-compressed (.tar.gz) or zip-compressed (.zip). It must contain one or more folders, each of which corresponds to a sample and must contain 3 files in it, (1) barcode.tsv[.gz], (2) features.tsv[.gz], and (3) matrix.mtx[.gz]). If the compressed file also contains “meta_data.csv” file in its top folder, it must have a header line and rows of the same number as the number of folders in the compressed file with the folder names in the 1st column. The pipeline requires “meta_data.csv” contains “sample” and “condition” columns to run DEG/GSEA. DEG/GSEA will not be performed if no “meta_data.csv” is provided or “meta_data.csv” does not contain both “sample” and “condition” column. If “sample“ column exists in the “meta_data.csv”, inference of cell-cell interaction will be performed per-sample basis and a summary of cell-cell interactions per-condition will be generated by collecting cell-cell interactions commonly found in at least half of the samples.

“cell-by-gene matrix” is a csv file (1st column being cell barcodes, 1st row being gene name (Hugo symbol). If the compressed input file contains “meta_data.csv” file as well, it must have the same number of rows as in the cell-by-gene matrix file, in which the 1st row is header line and the 1st column is the cell barcodes. The pipeline requires “meta_data.csv” contains “sample” and “condition” columns to run DEG/GSEA. DEG/GSEA will not be performed if no “meta_data.csv” is provided or “meta_data.csv” does not contain both “sample” and “condition” column. If “sample“ column exists in the “meta_data.csv”, inference of cell-cell interaction will be performed per-sample basis and a summary of cell-cell interactions per-condition will be generated by collecting cell-cell interactions commonly found in at leat half of the samples.