The research team, led by Mamta Masand and Harmeet S. Chawla, assembled the genome of HEAR genotype 13R4502 using PacBio HiFi long-read sequencing, a technology that reads stretches of DNA tens of thousands of bases long with high accuracy. The resulting assembly, named UM_B. napus Ver1, spans roughly 1.03 gigabases—close to the expected size of the rapeseed genome—with about 97 percent of the sequence anchored to the crop’s 19 chromosomes. That near-complete anchoring matters enormously for breeders, because it means almost every gene can be placed in its correct physical location, allowing variation between varieties to be compared chromosome by chromosome rather than scattered across disconnected fragments.

Quality control metrics for the assembly are striking. The benchmarking tool BUSCO, which checks for the presence of universally conserved single-copy genes, found 99.6 percent complete, indicating that the assembly captures essentially the full gene repertoire of the organism. The researchers projected 107,468 protein-coding gene models, a high number that reflects rapeseed’s famously complex genome: B. napus is an allotetraploid, formed by hybridization between the ancestors of turnip and cabbage, and it carries overlapping sets of genes inherited from both progenitor species. Repetitive elements accounted for just over 60 percent of the assembly, a reminder that the majority of a plant genome consists of transposable elements and other repeated sequences that have historically made assembly difficult.

It is precisely this repetitive, duplicated complexity that makes long-read sequencing so transformative. Earlier reference genomes, often built from short-read data, collapsed duplicated genes together or broke through repetitive regions, leaving gaps exactly where biologically important variation could hide. Because long reads span these difficult regions, the Manitoba team’s assembly can distinguish gene copies that short-read approaches would have merged, and it can reveal large structural changes—deletions, insertions, inversions and duplications—that smaller fragments of DNA simply cannot span. Structural variants are increasingly recognized as major drivers of trait differences in crops, even though they are far harder to detect than the single-letter changes that older genotyping methods focused on.

To hunt for such variation, the researchers did something methodologically clever: they compared reads from two different sequencing platforms. Alignments of PacBio HiFi reads were assessed independently against Oxford Nanopore reads, and only structural variants supported by both platforms were retained. This cross-platform filtering is a powerful defense against technology-specific artifacts, because errors that plague one chemistry may not occur in another. The analysis yielded 47,182 structural variants confirmed by both datasets—a remarkable catalog of large-scale genomic differences in this HEAR genotype relative to existing references.

Within that catalog, the team focused on structural variants overlapping genes involved in fatty acid metabolism, the biochemical pathway that determines how much erucic acid the seed accumulates. Sixteen fatty-acid-related candidate variants emerged, affecting members of gene families central to lipid biosynthesis and elongation. Among them were FAD2, which encodes a desaturase that converts oleic acid into linoleic acid, and FAE1, also known as KCS18, a beta-ketoacyl-CoA synthase that catalyzes the chain-elongation step responsible for producing very-long-chain fatty acids like erucic acid. Variants were also found in other KCS family genes, in KAR2 and in ACX family genes, which participate in fatty acid elongation and peroxisomal beta-oxidation. Because these enzymes sit directly on the road from simple fatty acids to erucic acid, structural changes within or near them are prime candidates for shaping the HEAR phenotype.

The researchers strengthened their evidence through additional layers of validation. Comparisons against multiple existing B. napus reference genomes provided independent support for four of the candidate variants, showing that the differences were not simply artifacts of aligning against a single reference. More decisively, polymerase chain reaction genotyping confirmed two of the candidates directly in the lab: a 66-base-pair deletion within FAD2 and a 51-base-pair insertion within FAE1/KCS18 in genotype 13R4502 relative to the widely used reference genome ZS11. PCR validation is a gold-standard check in this field, since it physically amplifies the region in question and shows that the size difference is real, not a computational artifact.

The findings are particularly interesting given what is already known about erucic acid genetics. FAE1 has long been established as a major gene controlling erucic acid content, and FAD2 is central to the opposite trait—the low-erucic, high-oleic oil that dominates edible canola markets. Finding structural variants in both genes in a high-erucic-acid genotype suggests that large-scale DNA rearrangements may complement the well-studied point mutations and copy-number changes known to influence these traits. If confirmed functionally, such variants could become precise molecular markers, letting breeders screen seedlings for the DNA differences rather than waiting years to measure oil composition in field-grown plants.

The authors are careful to note that discovery is not the same as proof. As they state, the functional contributions of these candidate variants to erucic acid accumulation require further transcriptomic, biochemical and genetic validation. Showing that a structural variant changes gene expression, alters enzyme function, or tracks consistently with oil phenotype across breeding populations would be the next steps. Nevertheless, the new assembly fills a genuine resource gap: despite several B. napus reference genomes already available, structural variation linked to erucic acid biosynthesis in HEAR germplasm had remained insufficiently characterized until now.

The broader implications reach beyond one Canadian genotype. UM_B. napus Ver1 is openly accessible, and its combination of contiguity, high completeness and validated variants makes it a template for future pan-genome work in rapeseed, where multiple high-quality assemblies from diverse lines can be compared to capture the full spectrum of genomic diversity. As industry demand for erucic acid grows alongside markets for biodegradable lubricants and bio-based polymers, understanding the structural genomic architecture that drives oil composition could accelerate the development of HEAR varieties with higher yields and tailored fatty acid profiles. What began as a technical exercise in long-read assembly may ultimately help turn a Winnipeg breeding program’s elite lines into the industrial oil crops of the future.

Subject of Research: Structural genomic variation in fatty acid metabolism genes of high-erucic-acid rapeseed revealed by long-read genome assembly

Article Title: Long-read genome assembly of high-erucic-acid Brassica napus reveals candidate structural variants in fatty acid metabolism genes

Article References: Long-read genome assembly of high-erucic-acid Brassica napus reveals candidate structural variants in fatty acid metabolism genes. (n.d.). https://doi.org/10.1186/s12864-026-13469-1

Image Credits: AI Generated

DOI: 10.1186/s12864-026-13469-1

Keywords: Brassica napus, high-erucic-acid rapeseed, genome assembly, PacBio HiFi, long-read sequencing, structural variants, fatty acid biosynthesis, FAE1, FAD2, erucic acid, plant genetics, BMC Genomics