remove_hydrogens_in_dir.py

import os

def process_pdb_files(directory):
    # Iterate through all files in the specified directory
    for filename in os.listdir(directory):
        if filename.endswith(".pdb"):  # Check if the file is a PDB file
            filepath = os.path.join(directory, filename)
            with open(filepath, 'r') as file:
                lines = file.readlines()

            # Process lines: skip HETATM and hydrogen ATOM lines
            #processed_lines = [line for line in lines if not line.startswith("HETATM") and not (line.startswith("ATOM") and line[76:78].strip() in ["H", "D"])]
            processed_lines = [line for line in lines if not (line.startswith("ATOM") and line[76:78].strip() in ["H", "D"])]

            # Write the processed lines back to the file
            with open(filepath, 'w') as file:
                file.writelines(processed_lines)
            print(f"Processed {filename}")

if __name__ == "__main__":
    directory = "./5j8h_exp/"  # Change this to your directory path
    process_pdb_files(directory)