What is Syrona? | How it differs from other OHDSI tools | Installation | Dependencies | Quick start | 1. Connect to a database | 2. Extract a dataset | 3. Compare two datasets | 4. Disconnect | Output structure | Next steps

Overview | Connecting to a database | PostgreSQL database | DuckDB (local files) | Running extraction | All domains (default) | Single domain | Shorthand: pass a DuckDB path directly | What each extractor does | Denominators (ACHILLES-116) | Condition prevalence (ACHILLES-404) | Condition chapters | Drug prevalence | Drug chapters | Procedure chapters | k-Anonymity | Loading and listing datasets | Disconnect

Overview | Defining a cohort | Care-site cohorts | Create cohorts | Extract by cohort | Verify a cohort | Delete a cohort | Cohorts from data frames | Care-site cohort options | Collapse strategy | Observation period clipping | How cohort filtering works | Compatibility with other OHDSI tools

Step 0 - Prerequisites | Finding your schemas | Step 1 - Open the SSH tunnel and connect | Verify | For local DuckDB users | Step 2 - Explore care sites | Step 3 - Create two cohorts | Step 4 - Extract the first cohort | Step 5 - Extract the second cohort | Step 6 - Compare | Step 7 - Launch the dashboard | Step 8 - Cleanup and disconnect | Troubleshooting | install_github() fails with "HTTP error 401 / Bad credentials" | "Could not connect" / "Connection refused" | "Permission denied for schema ..." | dbListTables() returns empty | "Cohort 1: 0 rows inserted" | extract_all() is very slow | Dashboard launches but dropdowns are empty | What to read next

Goal | Package versions used for the wiki page | Required environment variables | Connect to the OMOP database | Define the observation windows | Identify eligible ICD10 three-digit codes | Helper for clean study names | Build the target cohort | Build the visit-based control cohort | Keep only the final selected concepts | Run one ICD10 code and one observation window | Run the full atlas generation loop | Resulting output folders

Introduction | Installation | Load packages | Initiating database connection | Configuring Python dependencies (for GUI and summary precompute) | Air-gapped server setup | Building a target cohort | 1. Target cohort from OHDSI OMOP database. | 2. Target cohort from JSON description file. | 3. Target cohort from a CSV file. | 4. Target cohort from a table | Building a control cohort | 1. Control cohort based on matches | 2. Control cohort based on inverse controls | Other considerations

Goal | Expected JSON layout | Notes | Package versions used for the wiki page | Required environment variables | Connect to the OMOP database | Define study cases | Build the target cohort | Build the visit-based control cohort | Keep only the final selected concepts | Run one final mapped study | Run all configured cases | Resulting study folders

Overview | Running a comparison | All domains | Single domain | Step-by-step usage | Prevalence ratio formula | Output columns | Meta-analysis strategy | Heterogeneity metrics | Loading and listing comparisons

Scope | What differs in air-gapped environments | Offline setup workflow | 1. Prepare wheels on a connected machine | 2. Transfer wheels to the air-gapped server | 3. Configure Python in R | 4. Install dependencies from local wheels | 5. Validate before production usage | Offline smoke test | Summary precompute | GUI in server context | Runtime mode guidance for servers | Common failures and fixes | Minimal offline checklist

Introduction | What the composite shows | How to use it | Interpretation guidance

Introduction | Components | Controls | Patient vs Summary mode behavior | Interpretation

Executing the study | The parameters | Mandatory: | Customization: | Notes: | Reloading a saved study

Introduction | Select a study and mode | Main workflow in the UI | Tabs | Mappings overview

Introduction | Manual Merge | Hierarchy Suggestions | Correlation Suggestions | Mapping History

Introduction | Components | Interpretation

Introduction | How each mode is produced | What is different in the UI | Recommended usage

Introduction | Important model: staged vs applied | Action buttons | Apply Filters | Apply Table Selection | Recluster | Sidepanel controls | Heritage Types | Target Prevalence (%) | Prevalence Difference Ratio | Show ordinal data rows for active main concepts | Cluster Prevalence (%) | Top N Concepts by SD (across clusters) | Divergence Cluster Scope | Clusters | Clustering scope | Override and precedence rules | Mode-specific behavior | Patient mode | Summary mode | Recommended operating sequence

Introduction | Controls | Interpretation

Goal | Object Overview | data_initial | data_person | data_features | data_patients | complementaryMappingTable | selectedFeatureData | selectedFeatureData$selectedFeatureNames preview | selectedFeatureData$selectedFeatureIds preview | selectedFeatureData$selectedFeatures | conceptsData | conceptsData$concept_ancestor | conceptsData$concept | config | config$metadata scalar fields preview

Goal | Summary Folder Metadata (metadata.json) | concept_summaries.parquet | ordinal_summaries.parquet | clustering_k*_summary.parquet | clustering_k2_summary.parquet | clustering_k3_summary.parquet | clustering_k4_summary.parquet | clustering_k5_summary.parquet | clustering_k*_pairwise_overlap.parquet | clustering_k2_pairwise_overlap.parquet | clustering_k3_pairwise_overlap.parquet | clustering_k4_pairwise_overlap.parquet | clustering_k5_pairwise_overlap.parquet | complementaryMappingTable.parquet