added all ipynb files as well as figures for each methods.

Filtering_functions.py

@@ -0,0 +1,77 @@
+from rdkit import Chem
+from rdkit.Chem import rdMolDescriptors
+
+def atom_in_multiple_small_rings(mol):
+    """
+    Check if any atom is in two or more small (3- or 4-membered) rings.
+    
+    Args:
+        mol: RDKit molecule object.
+    
+    Returns:
+        bool: True if any atom is in two or more small rings, False otherwise.
+    """
+    # Get all ring information
+    ring_info = mol.GetRingInfo()
+    atom_rings = ring_info.AtomRings()  # Get the atom indices for each ring
+
+    # Create a dictionary to count how many rings each atom is part of
+    atom_ring_count = {}
+    for ring in atom_rings:
+        # Only count small rings (3- or 4-membered)
+        if len(ring) <= 4:
+            for atom_idx in ring:
+                if atom_idx in atom_ring_count:
+                    atom_ring_count[atom_idx] += 1
+                else:
+                    atom_ring_count[atom_idx] = 1
+
+    # Check if any atom is in more than one small ring
+    for count in atom_ring_count.values():
+        if count > 1:
+            return True  # Atom is in multiple small rings
+
+    return False  # No atom is in multiple small rings
+
+def double_bond_in_small_ring(mol):
+    """
+    Check if any small (3- or 4-membered) ring contains a double bond.
+    
+    Args:
+        mol: RDKit molecule object.
+    
+    Returns:
+        bool: True if a double bond exists inside a small ring, False otherwise.
+    """
+    # Get all ring information
+    ring_info = mol.GetRingInfo()
+    bond_rings = ring_info.BondRings()  # Get the bond indices for each ring
+
+    # Iterate through the rings
+    for ring in bond_rings:
+        if len(ring) <= 4:  # Only small rings (3- or 4-membered)
+            for bond_idx in ring:
+                bond = mol.GetBondWithIdx(bond_idx)
+                if bond.GetBondType() == Chem.rdchem.BondType.DOUBLE:
+                    return True  # Double bond found in a small ring
+
+    return False  # No double bonds in small rings
+
+def filter_molecules(mol):
+    """
+    Filter molecules based on the two rules:
+    1. No atom is part of two small rings.
+    2. No double bond inside small rings.
+    
+    Args:
+        molecules: List of RDKit molecule objects.
+    
+    Returns:
+        List of filtered molecule objects.
+    """
+    if not atom_in_multiple_small_rings(mol) and not double_bond_in_small_ring(mol):
+        return True
+    else:
+        return False
+
+

Molsnapper/Combine_sdf_files.py

@@ -0,0 +1,56 @@
+import os
+import shutil
+
+def get_unique_filename(dst_folder, filename):
+    """
+    Generates a unique filename by appending a counter if the file already exists in the destination folder.
+    
+    Args:
+        dst_folder (str): The destination folder.
+        filename (str): The original filename.
+    
+    Returns:
+        str: A unique filename.
+    """
+    base, extension = os.path.splitext(filename)
+    counter = 1
+    new_filename = filename
+
+    # Loop until we find a filename that does not exist
+    while os.path.exists(os.path.join(dst_folder, new_filename)):
+        new_filename = f"{base}_{counter}{extension}"
+        counter += 1
+
+    return new_filename
+
+def gather_shifted_sdfs(src_root, dst_folder):
+    """
+    Gather all *_shifted.sdf files from subfolders and copy them to a single destination folder.
+    Renames files with a counter if they already exist.
+    
+    Args:
+        src_root (str): The root folder containing subfolders.
+        dst_folder (str): The folder where all the *_shifted.sdf files will be copied.
+    """
+    if not os.path.exists(dst_folder):
+        os.makedirs(dst_folder)
+
+    # Walk through all directories and subdirectories
+    for root, dirs, files in os.walk(src_root):
+        for file in files:
+            if file.endswith('_shifted.sdf'):
+                # Get the unique filename to avoid overwriting
+                unique_filename = get_unique_filename(dst_folder, file)
+
+                # Build the full path to the source file
+                full_file_path = os.path.join(root, file)
+
+                # Copy the file to the destination folder with the unique filename
+                shutil.copy(full_file_path, os.path.join(dst_folder, unique_filename))
+                print(f"Copied: {full_file_path} to {os.path.join(dst_folder, unique_filename)}")
+
+if __name__ == "__main__":
+    src_root = '/home/yang2531/Documents/Bo_toolbox/PatWalters/Benchmarking_gene_model/Molsnapper/sample_MolDiff_20241002_214910_clash_rate_0.1_SDF'  # Change this to your parent folder
+    dst_folder = '/home/yang2531/Documents/Bo_toolbox/PatWalters/Benchmarking_gene_model/Molsnapper/sample_MolDiff_20241002_214910_clash_rate_0.1_SDF/combined_shifted_sdf'  # Change this to your destination folder
+
+    gather_shifted_sdfs(src_root, dst_folder)