**1. Introduction**

Rapeseed includes the Brassica oilseed crops in which *Brassica napus* is the leading crop globally, while *Brassica juncea*, *Brassica rapa*, *Brassica carinata*, and *Brassica nigra* are cultivated in selected regions of the world [1]. The term "canola" refers to the modified rapeseed low in erucic acid (<2%) and glucosinolate (<30 μmol/g of dried defatted meal) content, which includes cultivars of *B. napus*, *B. rapa*, and *B. juncea* [2]. The allopolyploid oilseed plants were formed following interspecific hybridization events, for example, *B. napus* (AACC, 2n = 38) was formed by hybridization of *B. oleracea* (CC, 2n = 18) and *B. rapa* (AA, 2n = 20) ~7500 years ago [3]. Modern approaches in evolutionary analysis have attempted to consolidate the origin of *B. napus* [4,5]. The use of rapeseed oil in lamps has been recorded since 2000 BC in India and since the 13th century in Europe [2] with additional uses in making food and soap [6]. Rapeseed cultivars were introduced to Canada in 1936 [7] for lubricant production for war ships, and was first commercially grown in Australia in 1969 [2].

Intensive breeding programs in *B. napus* were initiated in many countries around 1970, including Australia [2]. Back-crossing with the low-erucic- acid line, Liho, brought successful release of the first low erucic-acid cultivars of *B. napus* and *B. rapa* in 1968 and 1971, respectively [7]. High oleic and low linolenic (HOLL, > 65% oleic acid and < 3% linolenic acid) rapeseed genotypes were created by chemical mutagenesis using ethyl methanesulfonate (EMS) [8,9], and HOLL varieties have been commercially grown in Canada since 2005 and in Australia since 2006 [10]. The emergence of genomic and genetic tools has facilitated breeding programs [11,12] leading to the release of multiple rapeseed/canola varieties such as high erucic acid rapeseed (HEAR) [13], for non-food purposes, and varieties for

food industries e.g., high lauric acid canola Laurical ™, Roundup Ready ®(Monsanto, Missouri, US), InVigorTM (BASF), with provisional herbicide tolerance (HT) and resistance against major pathogens. The parental lines for rapeseed breeding programs varied depending on the geographic location, with the progenitors contributing towards the three ecotypes of *B. napus*, spring type which is widely cultivated in Canada, Australia, and northern Europe, winter type *B. napus* which is predominant in Asia and the remaining area of Europe [14], and semi-winter type as the primary rapeseed in China [15]. The increasing human population and sustainable energy initiatives have led to a high demand of canola oil for the food industry, biofuel, and various industrial purposes [16], prompting investments for further improvement of *B. napus*, with emphasis on the major traits relating to oil yield and quality, meal quality, herbicide tolerance, biotic and abiotic stress tolerance [17,18]. Fast-pace development of genetic and gene technologies has facilitated the identification of those breeding targets as well as achieving significant advances in the improvement of canola genomics resources.

During the last 20 years, dramatic innovations in next-generation sequencing (NGS) technologies have led to a significant drop in sequencing cost and o ffered exciting opportunities to explore plant genomes [19]. Recent rapeseed genomes were assembled using Illumina high-throughput short read sequence technologies and third generation sequencing technologies, notably, single molecule real-time (SMRT) sequencing (PacBio) and Nanopore sequencing technologies (Oxford Nanopore Technologies, Oxford, UK) capable of generating long-read sequences [20,21]. Genome assembly in highly complex genomic regions frequently found in *Brassica* species can be greatly improved using optical mapping [22,23] and chromosome conformation capture (Hi-C) technologies [23].

High-throughput SNP marker identification and genotyping of *Brassica* species, achieved through NGS, allow e fficient identification of quantitative trait loci (QTL) and generation of high density genetic maps, which means there is a higher chance of detection of a candidate gene linked to a nearby SNP [24,25]. The SNP markers are useful for genome-wide association studies (GWAS), marker-assisted selection (MAS), genomic selection (GS), and germplasm identification [26]. For example, SNP markers have been widely deployed in mapping QTL conferring resistance to blackleg, Sclerotinia stem rot [27] and clubroot [28] in *Brassica*. In addition, genotyping can be implemented quickly using the *Brassica* 60K Illumina Infinium ™ array with the capability to genotype of 52,157 SNPs in *B. napus* [29]. High-throughput genotyping technologies allow large-scale genomic characterisation of the huge *Brassica* germplasm collections worldwide comprising > 74,000 *Brassica* accessions mainly stored in The Netherlands, Norway, Spain, UK, US, and Australia [30] to facilitate improvement and broaden the genetic diversity of *Brassica* varieties.

Limitations in gene pool diversity [31,32], and the time and laborious constraints in canola breeding can now be overcome by using genetic modification [32] and genome editing technologies whereby transgenes or mutations can be directly introduced into plants, providing additional ways to investigate gene functions in biological processes [33–36]. The most commonly applied genome editing tool in eukaryotes in the last 8 years, the clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR-associated protein (Cas) system [37,38], is gaining more interest compared to conventional genetic modification methods due to its high e fficiency in targeted nucleotide modification and generation of a transgene-free end-product [39]. In a CRISPR/Cas system, a single-guide RNA (sgRNA) binds to the Cas protein and directs it to one or more specific target sequences [40] marked by a short sequence, protospacer-adjacent motif (PAM), located at the 3' downstream position [37] and varying in accordance with the type of Cas enzyme utilized [40,41]. Recently, a platform for induction of mutagenesis in oilseed rape via the CRISPR/Cas system has been developed [42], facilitating early utilization of this tool in improving agronomic traits of *B. napus* e.g., plant height [43], silique development [44], pod shatter resistance [45], and flowering time [46]. This genome editing tool provides a faster route to interpret the relationship between genes and phenotypic traits controlled by complex genetic structure, biosynthetic pathways, and regulatory elements [26].

Development of breeding techniques and the genetics and genomics associated technologies described above have improved the quality and agronomic traits of canola varieties, which make the current canola varieties well-suited for expectations of growers, processing sectors, and final users [47,48]. The advent of technologies, including omics technologies, genetic transformation, and genome editing, have facilitated deeper understanding of the physiological and biochemical processes regulating the development of the major phenotypic traits of *Brassica* oilseed crops, which has brought tremendous improvements in the desirable traits of canola in the last decade [49,50]. This review provides a summary of their impressive contributions towards improvement and innovations of canola varieties.

### **2. Improvement and Innovations of Canola Varieties**

### *2.1. Higher Resolution for Rapeseed Genome Characterization*

Genomic data is an integral part in developing breeding strategies and exploiting genetic potential of germplasm. The revolution in sequencing technologies has accelerated genome research, where it used to be an obstacle for polyploid species [21]. Using a combination of NGS and Sanger sequencing technologies, the first genome of *B. napus* was sequenced, from which its evolutionary history was clarified with the linkages to its ancient ancestor [5]. Meanwhile, the combination of PacBio's SMRT and NGS short read sequencing technologies have been deployed to study the various forms and complexity of the gene transcripts responsible for another development in *B. rapa* (Chinese cabbage), providing comprehensive transcriptome data for better accuracy of genome annotation [51]. Version 3 of the *B. rapa* genome was assembled using combinations of SMRT, optical mapping (BioNano), and Hi-C technologies with up ~ 30-fold improvement compared to the previous versions [52]. Other more recent genome assemblies for the *Brassica* species are *B. olerecea* accessions JZS v2 [53], *B. nigra* [54], *B. napus* 'Darmor-bzh [55], and German winter *B. napus* Express 617 [56]. These *Brassica* reference genomes were sequenced and assembled long-read sequencing technologies using SMRT and Oxford Nanopore Technologies (MinION and higher throughput version of the MinION, PromethION) complemented with Illumina data and verified by optical mapping and/or Hi-C data [52,56–58]. Qiao et al. [59] have proven cytoplasmic genomes from mitochondria and cytoplasms would be a highly e ffective approach for phylogenetic and evolutionary studies on *B. napus* including employing this resource in improved breeding.
