Datasets:
| #!/bin/python | |
| import pandas as pd | |
| import numpy as np | |
| from rdkit import Chem | |
| from rdkit.Chem import Descriptors | |
| from utils import * | |
| raw_data = pd.read_csv('raw/PDB-query_rawdata.csv') | |
| print(f'length of data before processing: {len(raw_data)}') | |
| # fill rows (containing different ligands) corresponding with the same crystal | |
| df_filled = fill_nans_iteratively(raw_data) | |
| print(f'length of data after filling NaN values (length should be the same): {len(df_filled)}') | |
| # filter by molecular weight of ligands to remove cofactors, crystallization agents, and ions | |
| df_filtered = filter_by_molecular_weight(df_filled) | |
| print(f'length of data after removing cofactors, crystallization factors, or ions: {len(df_filtered)}') | |
| df_filtered.to_csv('processed/processed_data.csv') | |
| print(f'data cleaning successful, resulting csv written to: processed/processed_data.csv') | |