Lmod_run module

RunLmodCalc

1. Class to Run Local vibrational mode calculations.
2. Generate a csv file from extracted QM information. (any name can be given)
3. Generate the geometry and LModA csv files for each molecules

current_dir_path (str): path to the folder which has optimized compounds as separate folders dataset (str): csv file with the information of smile and Zinc_id (optional)

Source code in qm40_dataset_for_ml/Lmod_run.py

class RunLmodCalc:
    """
    1. Class to Run Local vibrational mode calculations.
    2. Generate a csv file from extracted QM information. (any name can be given)
    3. Generate the geometry and LModA csv files for each molecules

    Args:
    current_dir_path (str): path to the folder which has optimized compounds as separate folders
    dataset (str): csv file with the information of smile and Zinc_id (optional)

    """

    def __init__(self, current_dir_path, dataset=None):
        self.current_dir_path = current_dir_path
        current_path = os.getcwd()
        target_folder_path = os.path.join(current_path, "Lmod_geom_results")
        if not os.path.exists(target_folder_path):
            os.makedirs(target_folder_path)
        self.target_folder_path = target_folder_path
        self.dataset = dataset
        self.dataset_name = "QM_parameters.csv"

    # running g16 results and Lmod calculations to extract g16 info, coordinates and Lmod info
    def run_lmod_collect_g16_info(self) -> list:
        list_dir = [
            name
            for name in os.listdir(self.current_dir_path)
            if os.path.isdir(os.path.join(self.current_dir_path, name))
        ]
        qm_parameter_collector = []
        smiles = []
        img_freq = []
        for current_folder in list_dir:
            current_folder_path = os.path.join(self.current_dir_path, current_folder)

            if ut.incomplete_running_checker(current_folder_path, current_folder):
                if ut.imaginary_freq_checker(current_dir_path, current_dir):

                    if self.dataset is not None:
                        smile = ut.get_smile(self.dataset, current_folder)
                        smiles.append(smile)
                    ut.lmod_clac(
                        current_folder_path, current_folder, self.target_folder_path
                    )
                    g16 = gaussian_info_extractor.GaussianInfoExtractor(
                        current_folder_path, current_folder
                    )
                    imaginary_freq = g16.imaginary_freq_checker()

                    if imaginary_freq is not None:
                        img_freq.append(imaginary_freq)

                    parameters, coordinates = g16.making_csvs()
                    qm_parameter_collector.append(parameters)
                    coordinate_name = f"{current_folder_path}_xyz.csv"
                    coordinates.to_csv(coordinate_name)
                    os.chdir(self.target_folder_path)
                    source_path = os.path.join(current_folder_path, coordinate_name)
                    target_fol_path = os.path.join(
                        self.target_folder_path, current_folder
                    )
                    shutil.copy2(source_path, target_fol_path)
        return qm_parameter_collector, smiles, img_freq

    # generate main dataset
    def final_dataset_generator(self) -> None:
        qm_parameters, smiles, img_freq = self.run_lmod_collect_g16_info()
        print(smiles)
        column_names = [
            "Internal_E(0K)",
            "HOMO",
            "LUMO",
            "Polarizability",
            "spatial extent",
            "dipol_mom",
            "ZPE",
            "rot1",
            "rot2",
            "rot3",
            "Inter_E(298)",
            "Enthalpy",
            "Free_E",
            "CV",
            "Entropy",
        ]
        df_init = pd.DataFrame(qm_parameters, columns=column_names)

        if self.dataset is not None:
            smile_df = pd.DataFrame(smiles)
            df_init = pd.concat([smile_df[0], df_init], axis=1)
            df_init = df_init.rename(columns={0: "smile"})
            df_init.to_csv(self.dataset_name, index=False)
            print(f"Final dataset saved as: {self.dataset_name}")

        else:
            df_init.to_csv("QM_parameters.csv", index=False)
            print(f"Final dataset saved as: {self.dataset_name}")

        if len(img_freq) != 0:
            df_img_freq = pd.DataFrame(img_freq, columns="imaginary_freq")
            df_img_freq.to_csv("imaginary_frequency_detect.csv")