Omics Data Preprocessing

Last updated: 2026-01-31

• Research project to identify spoilage factors in food using omics data
• Collaborated with Ewha Womans University BRL Lab
• My role: Data analysis and modeling → Results fed back to lab for further experiments
• What I did
  • Data preprocessing
    • Integrated multi-source datasets (Flavor, Material, Sanitation)
    • Processed taxonomic data across 5 levels (Phylum → Genus)
    • Validated taxonomy against GTDB database
  • Labeling & optimization
    • Developed rule-based Fresh/Spoiled classification using TPC, CC, TBARS, pH
    • Optimized thresholds with Grid Search, Optuna, Bayesian Optimization
  • Visualization & analysis
    • PCA biplot (CLR-transformed)
    • Relative abundance heatmaps
    • Identified candidate spoilage-related microbes for lab verification