COCONUT Database Data Preparation Summary
==========================================

Processing Date: 2025-12-12 22:02:57

Data Processing Pipeline:
1. Initial dataset: 715,822 compounds
2. After bacterial filtering: 24,911 compounds (3.48%)
3. After subsetting: 24,911 compounds
4. After SMILES validation: 24,910 compounds (100.00% valid)
5. After deduplication: 24,910 unique compounds

Summary Statistics:
- Compounds removed (invalid SMILES): 1 (0.00%)
- Duplicates removed: 0 (0.00%)
- Final dataset retention: 3.48% of original

Bacterial Filter Keywords Used:
bacteria, streptomyces, bacillus, actinobacteria, bacterial, prokaryot, mycobacterium, pseudomonas, clostridium, staphylococcus, escherichia

Organism Columns Searched:
organisms

Output Files:
- Raw data: /app/sandbox/session_20251212_135154_dad1cf0d70ee/data/coconut_csv-12-2025.csv
- Cleaned data: /app/sandbox/session_20251212_135154_dad1cf0d70ee/data/coconut_cleaned.csv
- This summary: /app/sandbox/session_20251212_135154_dad1cf0d70ee/results/data_prep_summary.txt

Data Quality Metrics:
- SMILES validation rate: 100.00%
- Deduplication rate: 0.00%
- Final unique bacterial compounds: 24,910

Success Criteria Met:
✓ data/coconut_cleaned.csv created with valid SMILES
✓ Dataset filtered for bacterial sources
✓ Data quality validation and deduplication completed