Recombination between heterologous human acrocentric chromosomes

Abstract

The short arms of the human acrocentric chromosomes 13, 14, 15, 21, and 22 share large homologous regions, including the ribosomal DNA repeats and extended segmental duplications. While the complete assembly of these regions in the Telomere-to-Telomere consortium's (T2T) CHM13 provided a model of their homology, it remained unclear if these patterns were ancestral or maintained by ongoing recombination exchange. Here, we use pangenomic resources and methods to show that specific pseudo-homologous regions of the acrocentrics recombine. Considering an all-to-all comparison of the high-quality human pangenome from the Human Pangenome Reference Consortium (HPRC), we find that contigs from all of the acrocentric short arms form a community similar to those formed by single chromosomes or the sex chromosome pair. A variation graph constructed from centromere-spanning acrocentric contigs indicates the presence of pseudo-homologous regions where most contigs appear as recombinants of heterologous CHM13 acrocentrics. A diploid T2T assembly of a target sample cross-validates these patterns. On chromosomes 13, 14, 21, and 22, we observe lower levels of linkage disequilibrium in pseudo-homologous regions than in their short and long arms, indicating higher rates of recombination and/or a larger effective population size. The ubiquity of signals of heterologous recombination seen in the HPRC draft pangenome's acrocentric assemblies suggests that this phenomenon is a basic feature of human cellular biology, providing sequence and population-based confirmation of hypotheses first developed cytogenetically fifty years ago.

Pangenome Graph Construction from Genome Alignment with Minigraph-Cactus

“…Pangenomes from Minigraph-Cactus cannot be used, for instance, to study centromeres. The omission of interchromosomal events will likewise preclude useful cancer pangenomes or studies into acrocentric chromosome evolution (Guarracino et al, 2022). We are also interested in ways to remove the necessity of filtering the graph to get optimal mapping performance by using an online method at mapping time to identify a subgraph that most closely relates to the reads of a given sample.…”

Long-read genomes reveal pangenomic variation underlying yeast phenotypic diversity

“…For SGE1 , three distinct alleles seen on chromosome XVI are also seen on other chromosomes, implying three independent recombination events, further underscoring the frequency of interchromosomal recombination. Interestingly, subtelomeric interchromosomal recombination has also been observed in humans (Guarracino et al 2022), and thus could contribute to human trait variation.…”

Building pangenome graphs

“…PGGB is implemented as a modular shell script, integrating independent components via standard text-based file formats, and in this provides a template for future pangenome construction methods. The method is practical, scalable to hundreds of genomes, and has been proven accurate through years of development in the Human Pangenome Reference Consortium (HPRC) [12,16] and in the broader bioinformatics community [17][18][19]. Here, we describe the specific innovations in the three main phases of the algorithm-alignment, graph creation, and graph normalization-which unlock this result.…”

“…pangenomes [16]. However, the use of WFMASH is not required, and PGGB supports the use of any user-defined input alignment set in PAF format.…”

“…PGGB has been applied and validated at large scale in projects in the HPRC [12], where it additionally has provided the first sequence-based evidence for systematic recombination between heterologous acrocentric human chromosomes [16]. Here, we present results from its application to a variety of diverse pangenome and comparative genomic contexts (Table 24).…”