split_bdg_HM.sh

#!/bin/bash

#A minimal shell script to split the bedgraph with H/M info into H and M bedgraghs

if [ $# -ne 1 ]; then
    echo "Usage: $0 <input_file.tsv>"
    exit 1
fi

#input file and two H/M files
INPUT_FILE=$1
H_OUTPUT="${INPUT_FILE%.tsv}_h.tsv"
M_OUTPUT="${INPUT_FILE%.tsv}_m.tsv"

# Process the file line by line
line_num=0
while IFS= read -r line; do
    # First line is the header - write to both files
    if [ $line_num -eq 0 ]; then
        echo "$line" > "$H_OUTPUT"
        echo "$line" > "$M_OUTPUT"
    else
        # Extract the 4th field (mod column)
        mod_value=$(echo "$line" | awk -F'\t' '{print $4}')
        
        # Send to appropriate output file
        if [ "$mod_value" = "h" ]; then
            echo "$line" >> "$H_OUTPUT"
        elif [ "$mod_value" = "m" ]; then
            echo "$line" >> "$M_OUTPUT"
        fi
    fi
    
    line_num=$((line_num + 1))
done < "$INPUT_FILE"

echo "Done!"
echo "H output: $H_OUTPUT"
echo "M output: $M_OUTPUT"