Once a candidate locus has been identified as interesting from a GWAS, haplotypes can be helpful to fine-map association signals. Further, understanding specific haplotypes that are most significant can help design sequencing studies and direct investigators towards underlying susceptibility variants. HapConstructor is implemented in a Monte Carlo framework for validity in family-based data. The algorithm automatically and iteratively builds and tests multi-locus SNP sets in a case-control framework. It begins with single SNPs, and expands these in a search to find SNP-haplotypes that optimize the association signal. Multi-SNP sets considered at any step in the process need not be contiguous; the SNP sets are built based on the significance of the preceding steps’ SNP subsets. Missing data imputation is included. Gene x Gene combinations and empirical false discovery rate thresholds are also provided. HapConstructor is a useful tool for exploring multi-locus associations in candidate genes and regions in a valid and structured process.
We note that when the design is *not* family-based data, the Monte Carlo framework may not be necessary. We provide hapcLite, a version of hapConstructor which uses standard distribution theory to assess significance within the iterative haplotype building.