An efficient multi-core package to reconstruct an underlying network of genomic signatures of high-dimensional epistatic selection from partially observed genotype data. The phenotype that we consider is viability. The network captures the conditional dependent short- and long-range linkage disequilibrium structure of genomes and thus reveals aberrant marker-marker associations that are due to epistatic selection. We target on high-dimensional genotype data where number of variables (markers) is larger than number of sample sizes (p >> n). The computations is memory-optimized using the sparse matrix output.