Genome-wide Imputation Using the practical haplotype graph in the heterozygous crop cassava

Genomic applications such as genomic selection and genome-wide association have become increasingly common since the advent of genome sequencing. Genotype imputation makes it possible to infer whole genome information from limited input data, making large sampling for genomic applications more feasible, especially in non-model species where resources are less abundant. Imputation becomes increasingly difficult in heterozygous species where haplotypes must be phased. The Practical Haplotype Graph is a recently developed tool that can accurately impute genotypes, using a reference panel of haplotypes.

The Practical Haplotype Graph is a haplotype database that implements a trellis graph to predict haplotypes using minimal input data. Genotyping information is aligned to the database and missing haplotypes are predicted from the most likely path through the graph. We showcase the ability of the Practical Haplotype Graph to impute genomic information in the highly heterozygous crop cassava (Manihot esculenta). Accurately phased haplotypes were sampled from runs of homozygosity across a diverse panel of individuals to populate the graph, which proved more accurate than relying on computational phasing methods.

At 1X input sequence coverage, the Practical Haplotype Graph achieves a high concordance between predicted and true genotypes (R=0.84), as compared to the standard imputation tool Beagle (R=0.69). This improved accuracy was especially visible in the prediction of rare and heterozygous alleles. We validate the Practical Haplotype Graph as an accurate imputation tool in the heterozygous crop cassava, showing its potential for application in heterozygous species.