Haplotype-Resolved Assembly in Polyploid Plants: Methods, Challenges, and Implications for Evolutionary and Breeding Research

Abstract

Polyploidization has been one of the key drivers of plant evolution, profoundly influencing plant adaptation in nature and crop traits in agriculture. Deciphering polyploid genomes is a crucial step for understanding evolutionary history and advancing agricultural applications. However, the inherent complexity of polyploid genomes has long hindered accurate assembly and annotation. Recent advances in sequencing technologies and improved assembly algorithms have significantly enhanced the resolution of complex polyploid genomes. These innovations have led to the successful assembly and public release of an increasing number of high-quality polyploid plant genomes. This review summarizes the mechanisms of polyploid formation and their evolutionary relevance, with a focus on recent technological progress in sequencing and genome assembly. On this basis, we further discuss the current key challenges of polyploid genome assembly and the ways to address them.

Keywords: crops; evolution; genome assembly; polyploidy.

Figure 1

Timeline of Arabidopsis (col−0 and polyploids) genome assembly advancements.

Figure 2

Genome assembly procedure of Overlap Layout Consensus (OLC) algorithm and de Bruijn graph (DBG); sequencing error bases are represented by red. (a) Overview of OLC algorithm. (b) Overview of de Bruijn graph algorithm. The main assembly path is represented by red arrow.

Figure 3

Three strategies for chromosome clustering in polyploid genome assembly. Arrows A, B, and C represent the traditional method, the ALLHiC method, and the HapHiC method, respectively. (a) Homologous chromosomes in a polyploid genome, where red regions indicate highly similar homologous segments. (b) Hi-C links among overlapping contigs: red lines represent collapse contigs; black dashed lines indicate allele Hi-C links; red dashed lines indicate Hi-C links between collapse and uncollapse contigs. (c) Traditional chromosome anchoring approach. (d) Hi-C link pruning based on allele tables from closely related species, retaining only the strongest Hi-C signals between collapse and uncollapse contigs. (e) Clustering after pruning. (f) Preprocessing of contigs, including the removal of low-information sequences, collapse contigs, and inter-allele Hi-C links. (g) Markov clustering on processed contigs. (h) Reassignment of remaining contigs.