Data history

1. As a first step towards building my dataset, I uploaded a CSV file downloaded from ChEMBL. In this CSV file, there are molecules with inhibitory activity against CDK5. The resulting molecules are 1049 (1048 rows × 46 columns) , and from here, I begin my work in constructing the dataset.
2. I selected, within a pandas dataset, the columns of interest, in this case, "Molecule ChEMBL ID" along with their corresponding "Standard Value" and "Smiles", resulting in a total of 1048 rows x 3 columns.
3. In this dataset, there are many repeated molecules that have different standard values. I'll search within the "Molecule ChEMBL ID" column for all molecules that are repeated more than once, resulting in a total of 28 molecule.
4. Since for each identical molecule, there is a different standard value, I calculated the average for each repeated molecule. I reformulated the dataset by removing duplicate molecules but keeping only the first occurrence, resulting in a total of 1019 rows × 3 columns.
5. When I calculated the average, I replaced the values within the dataset with the calculated averages, resulting in only one molecule (for those repeated) with its corresponding average, plus the rest of the molecules that were not repeated.