Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The Steering Group for the Multinational Brassica Genome Project published a concept note in 2003 for the first Brassica Genome Sequencing Project (http://brassica.nbi.ac.uk/brassica_genome_sequencing_concept.htm). B. rapa was selected first as it has the smallest genome among the cultivated Brassica species and fewer transposon-related sequences are interspersed between the genes than are found in B. oleracea, for example (Town et al. 2006; Yang et al. 2006; Cheung et al. 2009). The project aimed initially to produce, from bacterial artificial chromosome (BAC) clones, “Phase 2” sequence (i.e. fully oriented and ordered sequence but some small sequence gaps and low quality sequences) for the gene space of the ca. 500 Mb genome of B. rapa subspecies pekinensis, cultivar Chiifu. The activity was named the “Brassica rapa Genome Sequencing Project” (BrGSP).

Early activities of the BrGSP Consortium were centered on establishing mechanisms for information exchange, agreeing upon on the mapping populations to be used for anchoring sequences to genetic linkage maps and agreeing upon on the BAC libraries to be used for sequencing. Much of the early activities were led by the groups in South Korea, where major genomics research programs in B. rapa were already underway. Two mapping populations were agreed upon; both derived from the crosses between the cultivars Chiifu and Kenshin (CKDH and CKRI). Two BAC libraries were selected initially: KBrH and KBrB, both constructed in South Korea. Each library consists of 144 × 384-well plates; made using HindIII (KBrH) or BamHI (KBrB) digested genomic DNA. In all, approximately 20-fold redundant representation of the genome was made available.

The initial sequencing strategy was defined as a BAC-by-BAC approach, starting from seed BACs anchored genetically with extension based on overlaps between clones identified using BAC-end sequences. Several countries with major research programs involving Brassica species participated in the sequencing. Initially, the BAC libraries were end-sequenced (by groups in South Korea, Canada, UK, Australia and Germany), seed BACs were sequenced and mapped (largely by groups in South Korea), and the task of sequencing the genome was allocated on a chromosome-by-chromosome basis (involving groups in South Korea, UK, China, Canada and Australia). Later on, a complete BAC-based physical map was constructed (Mun et al. 2008) to improve the rate of progress.

The BAC-by-BAC approach to genome sequencing was based on capillary sequencing technology. Over 1000 BAC clones were sequenced, annotated and placed rapidly in the public domain, underpinning early insights in the sequence-level structure of Brassica genomes (Mun et al. 2009). However, several countries failed to fund the sequencing of chromosomes allocated to them as part of the BrGSP, so only chromosome A03 was completed by the strategy (Mun et al. 2010).

By 2009, advances in sequencing technology made strategies for sequencing complete genomes based on capillary sequencing obsolete. Brassica species had seemed unpromising subjects for the deployment of “Next Generation Sequencing” (NGS) technologies, which produced massively parallel but relatively short sequence reads, as extensive triplication had long been evident in Brassica genomes (O’neill and Bancroft 2000), potentially confounding assembly. However, increases in sequence read length and improvements in computational strategies overcame this potential barrier. In China, a whole-genome NGS approach was taken up in the Chinese Initiative of B. rapa sequencing, which involved sequencing of the genome of B. rapa cv. Chiifu, and produced excellent results. The BrGSP Consortium agreed in 2009 to abandon the BAC-by-BAC approach, focusing efforts on using insights from the higher-quality data to optimize the NGS-based approach and analysis (http://brassica.nbi.ac.uk/pdf/BrGSP_aug_2009.pdf).

The Chinese initiative assembled the B. rapa genome by SOAP-denovo (Li et al. 2010). They generated seven libraries with insertion size ranging from 184 bp to 10 Kb (Table 2.1). Three libraries ranging from 184 to 500 bp were used to assemble contigs while fourlibraries with large inserts ranging from 2 to 10 Kb were used to link the contigs to scaffords. To make full use of the existing resources and complement the disadvantage of the limited insertion size of Illumina sequencing DNA libraries, the Chinese initiative adopted a strategy combining the Illumina GAII data with BAC sequence data generated by the BrGSP. The assembly achieved by the Chinese initiative has an N50 contig size over 27 Kb and scaffold size over 339 Kb. Combining the assembled contigs from the Illumina GAII data with BAC sequence data, it has been produced 39 super-scaffolds with an N50 of over 1.97 Mb (Table 2.2). Excluding the highly abundant satellite sequences, the assembled sequence accounted for 284 Mb, of which 255 Mb (~90 % of the 284 Mb) has been anchored onto the ten chromosomes, covering 58.5 % of the estimated 485 Mb genome and about 98 % of the gene space. The number of predicted gene models for B. rapa is 41,174, about half as much again as Arabidopsis (Table 2.3).

Table 2.1 Summary of Illumina sequencing data for B. rapa genome
Full size table
Table 2.2 Summary of the final assembly statistics
Full size table
Table 2.3 General statistics for predicted protein-coding genes
Full size table

Because the B. rapa var. Chiifu chromosome A03 assembly (BAC A03) reportedby Mun et al. (2010) was completely based on the sequence data generated from traditional Sanger sequencer, it provided a perfect reference for the evaluation of the quality of B. rapa genome assembly by whole-genome shotgun (WGS) based on NGS data. After the Chinese team released the WGS assembly, several teams performed the evaluation by comparing the two A03 assemblies. The comparison showed very high level of agreement between both the Sanger sequenced BAC-by-BAC approach and the WGS approach. There are only minor discrepancies between Sanger and the NGS data. The total sizes of WGS A03 and BAC A03 are approximately 31.72 and 32.70 Mb, respectively, with slightly more repeat sequences assembled using the BAC approach (9.82 Mb in BAC A03 and 5.68 Mb in WGS A03). There were more gaps observed in BAC A03 (1035/1,358,889 bp, number of gaps/total size of gaps) than in WGS A03 (858/844,319 bp). Forty-four obvious inversions (>1 kb) between the two assemblies were verified by mapping the paired-end reads. The depth of the mapped reads and gaps at the boundaries for 38 inversions supported the WGS assembly, and six inversions remained ambiguous (Fig. 2.1).

Fig. 2.1
figure 1

Mummer plot of pseudochromosome A03 from Mun et al. versus that from the WGS assembly

Full size image

Based on this assembly, two groups did extensive gene synteny analysis of it with Arabidopsis. Xiaowu Wang’s group developed a gene synteny analysis pipeline specifically adapted to the closely-related species for identifying the accurate syntenic genes between Brassica and Arabidopsis (Cheng et al. 2012). Mike Freeling’s group analyzed synteny using CoGe (Tang and Lyons 2012). Both the analyses confirmed that the genome of B. rapa had undergone genome triplication subsequent to the last genome duplication observed in the genome of A. thaliana (the α duplication). Moreover, they found that the extent of gene loss (fractionation) among triplicated genome segments varies, with one copy containing a greater proportion of genes expected to have been present in its ancestor (70 %) than the remaining two (46 and 36 %). With this, they proposed a “two-step” hypothesis for B. rapa genome evolution, whereby one hybridization between diploid species occurred, following which genome fractionation occurred for a period of time before hybridization with a further diploid species, after which fractionation proceeded on all three subgenomes (Wang et al. 2011; Cheng et al. 2012; Tang et al. 2012).

One of the important goals of sequencing the B. rapa genome was to explain the extreme plasticity of the morphological variations, which can be found in B. rapa and other Brassica crops (Teutonico and Osborn 1994; Gustafson et al. 2006; Wittkop et al. 2009; Liu et al. 2012). Three possible factors contributing to the rich morphological polymorphism in the species were identified. The first factor may be a general increase in nucleotide substitution rates. The relatively recent polyploidizations in B. rapa may also have contributed to accelerated evolution due to genomic instability and gene redundancy. The third factor is the expansion of auxin-related gene families, as auxin controls many plant growth and morphological developmental processes. B. rapa has also experienced striking amplification of the plant-specific TCP transcription factor gene family, important in the evolution and specification of plant morphology.

Brassica Genome Gateway (http://brassica.nbi.ac.uk/), Brassica.Info (www.brassica.info) and http://www.brassica-rapa.org were the three most important web-based genome database for Brassica community when the BAC-by-BAC sequencing project was being conducted. Brassica.Info and Brassica Genome Gateway kept on updating regularly the data of the BACs being sequenced by the BrGSP consortium. Brassica Genome Gateway provided further annotation data of the sequenced BACs. The web site, http://www.brassica-rapa.org, hosted by the National Institute of Agricultural Biotechnology (NIAB) provided also annotated BAC information, mapping data and the physical map, which were generated in South Korea. After the Chinese Initiative finished the NGS sequencing project, Institute of Vegetables and Flowers (IVF) set up the Brassica database (BRAD, http://brassicadb.org), which provided services of the complete annotated Brassica A genome sequence (Cheng et al. 2011). It marked the completion of the B. rapa Genome Sequencing Project.

FormalPara Important events of the BrGSP Consortium

  • Jan. 2000: A Brassica Session was separated from the Arabidopsis Workshop for the Plant and Animal Genome Meeting held at San Diego, CA, USA during … (provide web site).

  • Apr. 2002: During the 13th Crucifer Genetics Workshop, there was acceptance of the requirement for bringing together various national projects under the banner of “Multinational Brassica Genome Project” (MBGP).

  • Jun. 2003: Steering Group for Multinational Brassica Genome Project was established and announced “Concept note for the Brassica Genome Sequencing Project”. The project aimed initially to produce, from BAC clones, “Phase 2” sequence for the ca. 500 Mb genome of B. rapa subspecies pekinensis and planned to finish the sequencing of the genome by the end of 2007.

FormalPara Important events of the Chinese Initiative of B. rapa Sequencing

  • Oct. 2008: IVF and BGI initiated B. rapa genome sequencing by NGS. IVF signed an agreement with BGI and started B. rapa genome sequencing activities.

  • Jan. 2009: The Chinese initiative produced the first draft assembly of the B. rapa genome with purely Solexa reads and sent the results to BrGSP Consortium members. BrGSP Consortium decided to have a meeting with the Chinese initiative members.

  • Mar. 2009: BrGSP Consortium members had a meeting with the Chinese initiative members. It was reported that the Chinese Initiative will finish the assembling of the B. rapa genome before July 2009. BrGSP Consortium decided to evaluate the quality of the Chinese assembly when it is finished.

  • May. 2009: Oil Crop Research Institute (OCRI) decided to join the Chinese initiative.

  • Jul. 2009: Chinese initiative sent the assembly of Chromosome A02, A03, A08 and A09 to BrGSP Consortium for evaluation.

  • Aug. 2009: The Chinese Initiative reported the B. rapa genome assembly based NGS in the Brassica Genome Sequencing meeting in Saskatoon, Canada. During the meeting a decision was made to accept the B. rapa genome assembly of the Chinese Initiative as the reference for the Brassica research community. BrGSP Consortium abandoned the BAC-by-BAC sequencing activities.

  • Jul. 2010: The Brassica Database (BRAD, http://brassicadb.org/), hosted by IVF/CAAS and providing searching and downloading services of all B. rapa genome sequences, was online, indicating the release of B. rapa genome sequence to the public (Cheng et al. 2011).

  • Aug. 2011: The B. rapa genome was published as “The genome of the mesopolyploid crop species B. rapa” in Nature Genetics (Wang et al. 2011).