📋 Column Role Definition (Omics Standard)
- 🔵 Row ID: Compound unique identifier (1st column by default, only for index matching, NO numerical calculation)
- 🩶 Metadata: Annotation info (KEGG, HMDB, Chinese Name, RT, classification, localization... fully reserved, never involved in math processing)
- 🟢 Sample Data: Concentration/intensity columns, ONLY columns for all data processing, imputation & scaling
- 🔴 Discard: Redundant columns, completely excluded from all procedures
✅ Auto-clean enabled: All non-numeric text (ND, N/A, Not Detected, symbols) in Sample Data columns → auto converted to missing values.