We recently developed methods for haplotype classification using chromosome-scale assemblies. Here we expand on this concept using ~13-fold whole genome shotgun (WGS) raw reads. We employ k-mer based algorithms to identify variations and then define haplotypes using multiple clustering approaches. We leveraged the pangenome assemblies (Walkowiak et al., 2020) to increase the precision of our method and WGS reads for elite and ancestral genotypes. Using this strategy, we could identify haplotype blocks from raw reads similarly to Brinton et al., (2020) without the need of chromosome-scale assemblies and could track haplotype blocks selected by breeders. We used our method to detect introgressions at 50-kb resolution and found previously unreported alien introgressions. This approach has the potential to integrate additional wheat cultivars, landraces, or wild relatives into the analysis and expand the haplotype database. This knowledge can be used for haplotype-guided decision making in breeding programs.