SplitTSV

Here we discuss the use of the SplitTSV command, which separates the fragment files by cell type according to a reference file with mapped barcode - cell-type associations.

Usage

SplitTSV \
    '$sample_directory' \
    '$output_directory' \
    '$barcodes'

Input Format

Required arguments:
sample_directory # Path prefix that stores all datasets containing fragment files (e.g., ~/atac_seq_data/mouse)
output_directory # Path prefix to store all output split fragment files by dataset (e.g., ~/atac_seq_data/mouse_split)
barcodes # Barcode file in .csv format. 
           First column must include sample_id:barcode (e.g., GSM4594382:AAACGAAAGAACCCGA-1)
           Seventh column must include the associated cell_type name (e.g., Somatotrophs)

Output format

Datasets will be output in the output_directory folder, with the following named format:

~/atac_seq_data/mouse_split/{dataset_id}/{dataset_id}_{cell_type}.tsv.gz