FSL_CP_DataPrep

Preparing data for few-shot learning with Cell Painting Assay

How to use

Order for running scripts: From 1 to 12 (11 can be skipped since we no longer use jsonl files)

/csv folder with 'chemical_annotations.csv', 'processed_data_concat.csv'

/sql folder with 'chembl_31_sqlite/chembl_31.db', or any version of ChEMBL

Blank /jsonl, /temp and /output folder. The /jsonl and /output folder contain the final data

Also refer to the screenshot attached for the layout of the folder

Name		Name	Last commit message	Last commit date
Latest commit History 21 Commits
utils		utils
.gitignore		.gitignore
1.chembl-full-regression.py		1.chembl-full-regression.py
10.create_other_mol_features.ipynb		10.create_other_mol_features.ipynb
11.create_jsonl_data.ipynb		11.create_jsonl_data.ipynb
12.create_final_label_df.ipynb		12.create_final_label_df.ipynb
2.chembl-full-classification.py		2.chembl-full-classification.py
3.create_cellpainting_index.ipynb		3.create_cellpainting_index.ipynb
4.build_label_matrix.ipynb		4.build_label_matrix.ipynb
5.process_duplicate_assay.ipynb		5.process_duplicate_assay.ipynb
6.extract_assay_target.ipynb		6.extract_assay_target.ipynb
7.data_split.ipynb		7.data_split.ipynb
8.create_CP_feature_df.ipynb		8.create_CP_feature_df.ipynb
9.image_processing.py		9.image_processing.py
LICENSE		LICENSE
README.md		README.md
Screenshot_repository.png		Screenshot_repository.png