Modification of RHtyper for WES data by adding machine learning. The WES-based RHtyper algorithm consists of 4 main steps: (1) variant profiling of SNPs/indels and coverage alterations. (2) Predicting RHD zygosity and hybrid alleles, RHCE∗C and RHCE∗c, and the zygosity of RHD c.1136C>T and RHCE c.48G>C using machine learning models. (3) Refining hybrid allele and breakpoint predictions using segmentation. (4) Generating likelihood scores using genotypes and phased haplotype likelihoods to rank candidate allele pairs. Finally, the candidate allele pair with the highest likelihood scores is considered as the predicted genotype. BAM, binary alignment map; QC, quality control.