Abstract
Motivation With the availability of new sequencing technologies, the generation of haplotype-resolved genome assemblies up to chromosome scale has become feasible. These assemblies capture the complete genetic information of both parental haplotypes, increase structural variant (SV) calling sensitivity and enable direct genotyping and phasing of SVs. Yet, existing SV callers are designed for haploid genome assemblies only, do not support genotyping or detect only a limited set of SV classes.
Results We introduce our method SVIM-asm for the detection and genotyping of six common classes of SVs from haploid and diploid genome assemblies. Compared against the only other existing SV caller for diploid assemblies, DipCall, SVIM-asm detects more SV classes and reached higher F1 scores for the detection of insertions and deletions on two recently published assemblies of the HG002 individual.
Availability and Implementation SVIM-asm has been implemented in Python and can be easily installed via bioconda. Its source code is available at github.com/eldariont/svim-asm.
Contact vingron{at}molgen.mpg.de
Supplementary information Supplementary data are available online.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Changed tone of introduction and discussion to avoid impression that SVIM-asm would be a replacement for read-based SV callers like pbsv; Updated evaluation results for genotyped SV calls; Transformed Figure 1 into stacked barplot of different error types; Updated Figures S5 and S6 to only show variants detected in chromosomal contigs