Comparing methods for constructing and representing human pangenome graphs

Abstract

Background: As a single reference genome cannot possibly represent all the variation present across human individuals, pangenome graphs have been introduced to incorporate population diversity within a wide range of genomic analyses. Several data structures have been proposed for representing collections of genomes as pangenomes, in particular graphs.

Results: In this work, we collect all publicly available high-quality human haplotypes and construct the largest human pangenome graphs to date, incorporating 52 individuals in addition to two synthetic references (CHM13 and GRCh38). We build variation graphs and de Bruijn graphs of this collection using five of the state-of-the-art tools: Bifrost, mdbg, Minigraph, Minigraph-Cactus and pggb. We examine differences in the way each of these tools represents variations between input sequences, both in terms of overall graph structure and representation of specific genetic loci.

Conclusion: This work sheds light on key differences between pangenome graph representations, informing end-users on how to select the most appropriate graph type for their application.

Keywords: Algorithms; Pangenomics; Sequence analysis; Variation graphs; de Bruijn graphs.

Fig. 1

The complete pangenome construction scheme and visualization. A The overall workflow, using 5 different tools on 3 different datasets; B complete 104 haplotypes variation graph built by Minigraph; C focus on part of HLA (MHC) region in chromosome 6 from panel B; D focus on DRB1-5 locus of HLA from panel C; E, complete 10 haplotypes variation graph built with pggb; F 10 haplotypes variation graph built with Minigraph-Cactus; G 104 haplotypes pangenome mdbg; H 10 haplotypes Bifrost dBG. All graphs except those produced by Minigraph have been simplified using gfatools and rendered using Bandage. VG is for variation graph

Fig. 2

Representations of the HLA-E locus by five graph construction methods over three increasing large human pangenomes. Nodes highlighted in red contain part of the locus sequence. The number of nodes and edges displayed below each graph concerns the whole subgraph (both red and gray nodes). Minigraph, on H2, H10, and H104, and mdbg, on H2, have only a portion of one node highlighted since the 4.8 kbp region is contained inside a single, long node

Fig. 3

Representations of the complex HLA region by five graph construction methods over three increasing large human pangenomes. See caption of Fig. 2 for details