The encyclopedia · R&D & Science · Technical decision · 2010–2011
VCF's eight fixed columns made variant data exchangeable across genomics
The 1000 Genomes Project's 2010 variant call format packed any variant into eight columns plus headers; it became genomics' exchange standard.
1000 Genomes Project
the move
Sequencing projects were producing growing lists of genetic variants — SNPs, insertions, deletions, structural changes — but each project stored them differently. The variant call format (VCF) was developed in 2010 for the 1000 Genomes Project as a generic format for storing DNA polymorphism data with rich annotations.
Its design choice was a fixed core with an extensible shell: eight mandatory tab-separated columns (chromosome, position, ID, reference allele, alternative alleles, quality, filter, info) cover every variant, while '##' header lines define optional INFO, FILTER and FORMAT fields. The format's authors designed it to scale to millions of sites across thousands of samples, with textual encoding plus indexing for fast access.
The project's weight carried the format: UK10K, dbSNP and the NHLBI Exome Project adopted it, and VCFtools shipped alongside it in 2011 to validate, merge and compare files. It is now the common output of variant-calling pipelines, with the specification maintained at version 4.5.
why it works
- Eight fixed columns cover the core of any variant
- Extensible INFO and FORMAT fields carry rich annotations
- Indexed text scales from one sample to thousands
- The 1000 Genomes Project's adoption made it the default for new projects
what transfers
The format that scales wins genomics: text for humans, indexable for machines, and extensible headers so the standard can grow without breaking old files.
what came after
VCF is the standard output of variant callers and the basis of gVCF for reference-matching blocks; BCFtools and VCFtools remain core utilities, and the hts-specs repository keeps the specification current.
references
spotted an error? The archive wants to know.