HaploVar

What is HaploVar

HaploVar defines local haplotypes and is designed to be used as part of a GWAS or genomic selection pipeline. HaploVar takes a VCF file and a LD matrix, calculates local haplotypes, identifies haplotype variants and formats the output to be compatible with a wide range of GWAS and genomic selection tools. Halotypes improve the ability of GWAS to detect QTLs and increase the trait prediction accuracy of genomic selection studies.

How to cite HaploVar

When using HaploVar, please cite the following publication.

MacNish, T. R., Al-Mamun, H. A., Bergmann, T., Bestry, M. S., Marsh, J. I., & Edwards, D. (2025). HaploVar: An R package for defining local haplotype variants for trait association and trait prediction analyses. Bioinformatics, 41(12), btaf602. https://doi.org/10.1093/bioinformatics/btaf602

Installation

HaploVar can be installed using the following code:

#Install the development version from github
install.packages("devtools")
# devtools::install_github("TessaMacNish/HaploVar")
install.packages("HaploVar", repos = c('https://tessamacnish.r-universe.dev', 'https://cloud.r-project.org'))

#or install directly from CRAN
install.packages("HaploVar")

library(HaploVar)

Documentaion

For a tutorial on how to use HaploVar please refer to introduction This tutorial describes how and when to use each function. Each parameter is explained, and example code is provided.

New Update - Nextflow Pipeline!

There is now a Nextflow pipline to help you run HaploVar. This pipeline will prepare the VCF for you and give you the final haplotype file. For example the pipeline will calculate the linkage disequilibrium matrices for you. Please access the Nextflow pipeline here.

HaploVar Workflow Overview

An example of a HaploVar workflow for preparing for genomic prediction is below:

  1. Collect your genotype data. This should be in VCF format and include SNP genotypes.
  2. Calculate linkage disequilibrium for your genotype data using a tool such as PLINK (Purcell et al. 2007, https://zzz.bwh.harvard.edu/plink/ld.shtml) or the ld function of snpStats (Clayton, 2025). For example:
plink --allow-extra-chr --r2 square --vcf vcf --out output.ld
  1. Read your VCF and linkage disequilibrium matrix into R. crosshap has an inbuilt function for this purpose (Marsh et al. 2023).
install.packages("crosshap")
library(crosshap)
VCF <- read_vcf("vcf")
LD <- read_LD("output.ld",vcf = VCF)
  1. Calculate haplotypes. The following code outputs haplotype tables displaying which SNPs group together to form haplotypes.
haplotype_list <- define_haplotypes(VCF, LD, epsilon = 0.8)
  1. Identify haplotype variants and format the output for genomic selection pipelines.
format3 <- haplotype_variants(VCF, LD, epsilon = 0.8, format = 3)
  1. Save the output
write.csv(format3, "output.csv", row.names=TRUE, quote = FALSE )
  1. Run your preferred genomic selection pipeline

NOTE: If you want to prepare haplotype variants for GWAS instead of genomic selection change format to equal 6 and save the data with the following code:

write.table(format6, "output.vcf", sep = "\t", row.names = FALSE, col.names = TRUE, quote = FALSE)

References

Clayton D. snpStats: SnpMatrix and XSnpMatrix classes and methods. R pack-age version 1.59.2. 2025. https://bioconductor.org/packages/snpStats

Marsh JI, Petereit J, Johnston BA et al. crosshap: R package for local haplotype visualization for trait association analysis. Bioinformatics 2023;39. https://doi.org/10.1093/bioinformatics/btad518

Purcell S, Neale B, Todd-Brown K et al. PLINK: A tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 2007;81:559–575. https://doi.org/10.1086/519795