LAVA is a tool to conduct genome-wide, local genetic correlation analysis on multiple traits, using GWAS summary statistics as input.

The primary publication for LAVA (currently under submission) is:

Werme J, van der Sluis S, Posthuma D, de Leeuw C (2021). LAVA: An integrated framework for local genetic correlation analysis. BiorXiv preprint. doi: 10.1101/2020.12.31.424652 (link)

When using LAVA, please refer to this paper.

The current R-implementation for LAVA, as well as documentation and tutorial, can be found in the GitHub repository here.

For questions, comments, feedback and feature requests, contact Josefin Werme.

Auxiliary files

The genome partition file used for the primary LAVA publication can be found in the support_data folder on the GitHub repository linked above, and the code used to create the partitioning can be found here. This partioning was created using the 1,000 Genomes phase 3 reference data (European panel, build GRCh37/hg19).

Ready to use PLINK format data files for the 1,000 Genomes phase 3 reference data are linked below for different ancestries. Note that when analysing data from a non-European ancestry, a new genome partitioning will need to be created based on the reference data for that ancestry using the code linked above.

PopulationFile SizeLink
European488 MBDownload
African1057 MBDownload
East Asian440 MBDownload
South Asian521 MBDownload
Middle/South American492 MBDownload
GTEx v8 eQTL and sQTL input files

[coming soon]