A variant selection framework for genome graphs

Abstract

Motivation Variation graph representations are projected to either replace or supplement conventional single genome references due to their ability to capture population genetic diversity and reduce reference bias. Vast catalogues of genetic variants for many species now exist, and it is natural to ask which among these are crucial to circumvent reference bias during read mapping. Results In this work, we propose a novel mathematical framework for variant selection, by casting it in terms of minimizing variation graph size subject to preserving paths of length α with at most δ differences. This framework leads to a rich set of problems based on the types of variants [e.g. single nucleotide polymorphisms (SNPs), indels or structural variants (SVs)], and whether the goal is to minimize the number of positions at which variants are listed or to minimize the total number of variants listed. We classify the computational complexity of these problems and provide efficient algorithms along with their software implementation when feasible. We empirically evaluate the magnitude of graph reduction achieved in human chromosome variation graphs using multiple α and δ parameter values corresponding to short and long-read resequencing characteristics. When our algorithm is run with parameter settings amenable to long-read mapping (α = 10 kbp, δ = 1000), 99.99% SNPs and 73% SVs can be safely excluded from human chromosome 1 variation graph. The graph size reduction can benefit downstream pan-genome analysis. Availability and implementation : https://github.com/AT-CG/VF. Supplementary information Supplementary data are available at Bioinformatics online.

A hepatitis B virus (HBV) sequence variation graph improves alignment and sample-specific consensus sequence construction

“…We should note that effective tools enabling sequence-to-graph alignment and the subsequent identification of graph-derived genetic variation are a relatively recent development. Improvements in computational performance have already been demonstrated through graph simplification and the development of more advanced mapping and variant identification models [52][53][54], with further improvements expected [55].…”

Computational graph pangenomics: a tutorial on data structures and their applications

“…This means that analysis on pangenome graphs becomes orders of magnitude slower than on linear references, and the impact of such analysis needs to be assessed (Chen et al 2021). Recent research tries to ameliorate this shortcoming by focusing on variant selection approaches that aim to reduce the size of the pangenome graph and speed up mapping (Jain et al 2021). With the maturation of the field of computational pangenomics, it is expected that tools with better performance will be developed.…”

Haplotype-aware sequence alignment to pangenome graphs

“…The number of sequences spelled by a graph increases combinatorially with the number of variants. This issue has been addressed previously by using different techniques, e.g., by limiting the amount of variation in the graph [23,44,58], artificially simplifying complex regions [16], or restricting the alignment to either one [38,57,56] or two haplotype paths [4]. A more principled approach to tackle this problem may be to leverage the correlations between two or more genetic variants, i.e.…”