What Should I Run?

Use this page to choose the right bff-tools mode from your starting point. It is a decision page, not a tutorial.

If you want to run the bundled test data now, go to Quick Start. If you already know the mode and want copy-paste commands, use Command Recipes.

Decision Table

I have...	I want...	Run
XLSX metadata	Validate Beacon metadata and write BFF JSON collections	bff-tools validate
BFF JSON metadata	Validate existing JSON collections	bff-tools validate
A VCF or VCF.gz file	Generate BFF genomicVariations	bff-tools vcf
A SNP-array TSV/TXT file	Convert SNP-array data into BFF genomicVariations	bff-tools tsv
Existing BFF files	Load them into MongoDB	bff-tools load
Metadata plus VCF/TSV input	Convert genomic data and load everything into MongoDB	bff-tools full
Generated BFF output	Inspect it without MongoDB	bff-browser
BFF data already in MongoDB	Query it with a small web/API layer	bff-portal
Many repeated jobs	Queue and monitor them locally	bff-queue

How to Decide

Start with validate when

• your first input is a Beacon metadata workbook
• you want to check existing BFF JSON metadata
• you need individuals.json, biosamples.json, runs.json, or datasets.json before loading

Start with vcf or tsv when

• your first input is genomic data
• you only need to generate genomicVariations
• MongoDB loading can wait until after you inspect the output

Use vcf for VCF or VCF.gz. Use tsv for SNP-array style TSV or TXT input.

Start with load when

• BFF metadata collections already exist
• genomic variation output already exists, if your Beacon includes genomic data
• MongoDB is configured and reachable

Use full when

• you already have a working parameter file
• metadata paths are configured
• MongoDB settings are configured
• you want conversion and loading in one run

For a first real dataset, running validate, then vcf or tsv, then load is easier to debug than starting with full.

Avoid These Mixups

Situation	Better choice
You have not tested the install yet	Run Quick Start first
You only want to see command syntax	Use Command Recipes
You need file and log locations	Read Outputs
You are reviewing a dataset for sharing	Read Validation and Reproducibility
You need the complete workflow explanation	Read Data Beaconization

Next Step

If you are new to the toolkit, run the Quick Start first. If you already know your mode, continue to Command Recipes, the CLI reference, or the data beaconization workflow.