Python Library: Chemical_Descriptors

This function generates one of several fingerprint types from the list of molecular fingerprints available, each serving specific tasks in cheminformatics and computational chemistry.

Importance of Fingerprint Types:

• Distinct Representation: Different fingerprint types capture various aspects of a molecule’s structure, allowing for versatile molecular comparisons.

• Diverse Applications: Depending on the task (such as similarity searching, classification, or clustering), choosing the right fingerprint type ensures better performance in chemical analysis and predictive modeling.

• Accuracy in Modeling: The right fingerprint type can significantly improve the accuracy of machine learning models and predictions based on molecular features.

Number of Fingerprints:

Functions

cal_rdkit_descriptor(input_file, output_file, smiles_column)

Description:
This function calculates RDKit molecular descriptors for molecules specified in a CSV file (input_file) using SMILES strings from a specified column (smiles_column). The calculated descriptors are appended as additional columns to the original data and saved in a new CSV file named <input_file_name>_rdkit_descriptor.csv.

Parameters:

• input_file (str): Path to the input CSV file containing molecular data in SMILES format.
• output_file (str): Path where the output CSV file will be saved (optional if using the default naming convention).
• smiles_column (str): The name of the column in the input CSV file that contains the SMILES strings.

Output:
The output will be saved as a CSV file named <input_file_name>_rdkit_descriptor.csv.

cal_lipinski_descriptors(file_path, smiles_column, verbose=False)

Description:
This function calculates Lipinski descriptors for molecules specified in a CSV file (file_path) using SMILES strings from a specified column (smiles_column). It automatically saves the calculated descriptors to an output file named <input_file_name>_lipinski_descriptors.csv.

Parameters:

• file_path (str): Path to the input CSV file containing molecular data in SMILES format.
• smiles_column (str): The name of the column in the input CSV file that contains the SMILES strings.
• verbose (bool, optional): If True, the function will print additional processing details. Default is False.

Output:
The output will be saved as a CSV file named <input_file_name>_lipinski_descriptors.csv.

cal_morgan_fpts(input_file, smiles_column)

Description:
Calculates Morgan fingerprints for molecules specified in a CSV file (input_file) using SMILES strings from a specified column (smiles_column). The function saves the calculated fingerprints to an output file named <input_file_name>_calculate_morgan_fpts.csv.

Parameters:

• input_file (str): Path to the input CSV file containing molecular data in SMILES format.
• smiles_column (str): The name of the column in the input CSV file that contains the SMILES strings.

Output:
The output will be saved as a CSV file named <input_file_name>_calculate_morgan_fpts.csv.

cal_mordred_descriptors(input_file, smiles_column)

Description:
Computes Mordred descriptors for molecules specified in a CSV file (input_file) using SMILES strings from a specified column (smiles_column). The function saves the computed descriptors to an output file named <input_file_name>_mordred_descriptors.csv.

Parameters:

• input_file (str): Path to the input CSV file containing molecular data in SMILES format.
• smiles_column (str): The name of the column in the input CSV file that contains the SMILES strings.

Output:
The output will be saved as a CSV file named <input_file_name>_mordred_descriptors.csv.

calculate_selected_fingerprints(input_file, smiles_column)

Description:
Before using this function, execute the following code snippet to download and unzip the necessary files:

! wget https://github.com/dataprofessor/padel/raw/main/fingerprints_xml.zip
! unzip fingerprints_xml.zip
## Molecular Fingerprint Calculation

This function calculates 12 different types of molecular fingerprints:

- `AtomPairs2DCount`
- `AtomPairs2D`
- `EState`
- `CDKextended`
- `CDK`
- `CDKgraphonly`
- `KlekotaRothCount`
- `KlekotaRoth`
- `MACCS`
- `PubChem`
- `SubstructureCount`
- `Substructure`

Each enhanced dataset with fingerprints is saved as separate CSV files, appended with the respective fingerprint type name.

### Parameters:
- `input_file` (str): Path to the input CSV file containing molecular data in SMILES format.
- `smiles_column` (str): The name of the column in the input CSV file that contains the SMILES strings.

### Output:
Each fingerprint type will be saved as a separate CSV file with the respective fingerprint type name appended.  
For example: `<input_file_name>_AtomPairs2DCount.csv`.

---

## **fps**(filename, smiles_column, fp_type)

**Description:**  
This function calculates a specified molecular fingerprint (`fp_type`) for each molecule in a CSV file. The user must provide:
- The CSV file (`filename`)
- The SMILES column name in the file (`smiles_column`)
- One of the following fingerprint types:

  - `maccs`
  - `avalon`
  - `pattern`
  - `layered`
  - `map4`
  - `secfp`
  - `erg`
  - `estate`
  - `avalon-count`
  - `ecfp`
  - `fcfp`
  - `topological`
  - `atompair`
  - `rdkit`
  - `ecfp-count`
  - `fcfp-count`
  - `topological-count`
  - `atompair-count`
  - `rdkit-count`

### Parameters:
- `filename` (str): Path to the input CSV file containing molecular data in SMILES format.
- `smiles_column` (str): The name of the column in the input CSV file that contains the SMILES strings.
- `fp_type` (str): The type of molecular fingerprint to calculate (choose from the list of fingerprint types above).

### Output:
The output will be saved as a CSV file named `<input_file_name>_<fp_type>.csv` depending on the fingerprint type chosen.


Ahmed Alhilal
=============

0.0.1 (05/01/2025)
-------------------
- First Release