Requirements | Install from CRAN | Install from GitHub | Verify the installation | Update | Troubleshooting | R cannot install a dependency | GitHub cannot be reached | R loads an older version | Uninstall | Where to begin | Getting help

Discover a palette | Compare it in Palette Lab | Learn from an R recipe | Work with the same data | Follow one complete path | 1. Read the palette | 2. Look at the graphical behavior | 3. Open the R recipe | 4. Inspect or download the data | 5. Use the palette in R | Choose where to begin

Find a palette | Retrieve colors | Use a discrete scale | Use a continuous gradient | Preview one palette | Convert color formats | Next steps

Browse all palettes | Qualitative palettes | Sequential palettes | Diverging palettes | Source labels and new mappings | Use the selected palette | Explore further | Propose a palette

Overview | Prerequisites | Exploring Remote Files | List files and folders | Browse the directory tree | Generating Download URLs | Downloading Files | Single file or folder | Metadata shortcuts | Common Options | Getting Help

Purpose | Supported Sources | Default Reconciliation | Not Currently Supported | Design Principle | Related Articles

Welcome to ukbflow | Installation | A Quick Taste | Load data | Derive a disease phenotype | Run an association model | Plot the results | Full Function Overview | End-to-End Case Study | Additional Resources

Overview | Step 1: Validate the Weights File -- grs_check() | Step 2: Convert BGEN to PGEN -- grs_bgen2pgen() | Step 3: Calculate GRS Scores -- grs_score() | Step 4: Standardise GRS Columns -- grs_standardize() | Step 5: Validate Predictive Performance -- grs_validate() | Logistic (cross-sectional) | Cox (survival) | Complete Pipeline Example | Getting Help

1 Introduction | 2 Data Loading | 3 Decode Column Names | 4 Derive Phenotypes | 5 Exposure Definition | 6 Cohort Assembly | 7 Association Analysis | 8 Visualisation | Getting Help | 9 Session Info | 10 References

Overview | The Three-Model Framework | Step 1: Cox Proportional Hazards | Step 2: Logistic Regression | Step 3: Linear Regression | Step 4: Proportional Hazards Assumption Test | Step 5: Subgroup Analysis | Step 6: Dose-Response Trend Analysis | Step 7: Fine-Gray Competing Risks | Working with Results | Step 8: Lag Sensitivity Analysis | Getting Help

Overview | Start an Audit | Record Field IDs | Snapshot Data States | Record Phenotype Summaries | Record Cohort Assembly | Record Model Results | Review and Write the Manifest | Suggested Audit Points

Overview | Setup | Step 1: Handle Informative Missing Labels | Step 2: Prepare Covariates | Step 3: Bin Continuous Variables | Step 4: Self-Reported Disease | Step 5: HES Inpatient Records | Step 6: First Occurrence Fields | Step 7: Cancer Registry | Step 8: Death Registry | Step 9: Combine Sources with derive_icd10() | Step 10: Final Case Definition | Getting Help

Overview | Quick Install | From CRAN (recommended) | From GitHub (latest development version) | System Requirements | Dependencies | Core Dependencies | Analysis Dependencies | Visualization Dependencies | Optional Dependencies | Install dxpy (Local Mode Only) | Authentication Setup | Local → RAP | RAP → RAP | Verify Installation | Update ukbflow | From GitHub | From CRAN | Troubleshooting | dx not found | Token expired or session lost | Installation fails on Windows | Network / Firewall issues | Uninstall | Getting Help | Next Steps

Overview | Obtaining an API Token | Storing Your Token Securely | Logging In | Local → RAP | RAP → RAP | Checking Authentication Status | Listing Available Projects | Selecting a Project | Logging Out | Token Expiry | Full Local → RAP Workflow | Getting Help

Overview | Recommended Workflow | Step 1: Decode Values | What gets decoded | Step 2: Decode Names | Name conversion examples | Long names | Getting Help

Overview | ops_setup() — Environment Health Check | ops_toy() — Synthetic UKB Data | Cohort scenario | Forest scenario | Reproducibility | ops_na() — Missing Value Diagnostics | Controlling CLI output with threshold | Programmatic use | ops_snapshot() — Pipeline Checkpoints | Recording snapshots | Viewing the full history | Silent recording | Resetting history | Snapshot Helpers | ops_snapshot_cols() — column names at a checkpoint | ops_snapshot_diff() — compare two checkpoints | ops_snapshot_remove() — drop raw columns after deriving | ops_set_safe_cols() — register study-specific protected columns | ops_withdraw() — Exclude Withdrawn Participants | Typical Workflow | Getting Help

Overview | plot_forest() — Forest Plot | Minimal example | Building the input data frame from assoc_*() results | Key parameters | plot_tableone() — Baseline Characteristics Table | With SMD, custom labels, and export | Getting Help

Overview | Step 1: Classify Timing — Prevalent vs. Incident | Step 2: Age at Event | Step 3: Follow-Up Time | Prevalent cases receive NA follow-up time | Auto-detection of death and lost-to-follow-up columns | Full Survival-Ready Pipeline | Getting Help

Overview | Typical Workflow | Monitoring a Job | Check status | Wait for completion | Retrieving Results | Get the file path | Load into R | Browsing Job History | Getting Help

Overview | Prerequisites | Step 1: Browse Available Fields | Step 2: Extract Data | Recommended: extract_batch() | Instance type | Quick inspection: extract_pheno() | A Note on Column Names | Getting Help