Wang Bo, Chougule Kapeel, Jiao Yinping, Olson Andrew, Kumar Vivek, Gladman Nicholas, Huang Jian, Llaca Victor, Fengler Kevin, Wei Xuehong, Wang Liya, Wang Xiaofei, Regulski Michael, Drenkow Jorg, Gingeras Thomas, Hayes Chad, Armstrong J Scott, Huang Yinghua, Xin Zhanguo, Ware Doreen
Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, USA.
Texas Tech University, 1006 Canton Ave, Lubbock, TX 79409-2122, USA.
NAR Genom Bioinform. 2024 Aug 9;6(3):lqae097. doi: 10.1093/nargab/lqae097. eCollection 2024 Sep.
(L.) Moench is a significant grass crop globally, known for its genetic diversity. High quality genome sequences are needed to capture the diversity. We constructed high-quality, chromosome-level genome assemblies for two vital sorghum inbred lines, Tx2783 and RTx436. Through advanced single-molecule techniques, long-read sequencing and optical maps, we improved average sequence continuity 19-fold and 11-fold higher compared to existing Btx623 v3.0 reference genome and obtained 19 and 18 scaffolds (N50 of 25.6 and 14.4) for Tx2783 and RTx436, respectively. Our gene annotation efforts resulted in 29 612 protein-coding genes for the Tx2783 genome and 29 265 protein-coding genes for the RTx436 genome. Comparative analyses with 26 plant genomes which included 18 sorghum genomes and 8 outgroup species identified around 31 210 protein-coding gene families, with about 13 956 specific to sorghum. Using representative models from gene trees across the 18 sorghum genomes, a total of 72 579 pan-genes were identified, with 14% core, 60% softcore and 26% shell genes. We identified 99 genes in Tx2783 and 107 genes in RTx436 that showed functional enrichment specifically in binding and metabolic processes, as revealed by the GO enrichment Pearson Chi-Square test. We detected 36 potential large inversions in the comparison between the BTx623 Bionano map and the BTx623 v3.1 reference sequence. Strikingly, these inversions were notably absent when comparing Tx2783 or RTx436 with the BTx623 Bionano map. These inversion were mostly in the pericentromeric region which is known to have low complexity regions and harder to assemble and suggests the presence of potential artifacts in the public BTx623 reference assembly. Furthermore, in comparison to Tx2783, RTx436 exhibited 324 883 additional Single Nucleotide Polymorphisms (SNPs) and 16 506 more Insertions/Deletions (INDELs) when using BTx623 as the reference genome. We also characterized approximately 348 nucleotide-binding leucine-rich repeat (NLR) disease resistance genes in the two genomes. These high-quality genomes serve as valuable resources for discovering agronomic traits and structural variation studies.
高粱（L.）Moench是全球重要的禾本科作物，以其遗传多样性而闻名。需要高质量的基因组序列来捕捉这种多样性。我们为两个重要的高粱自交系Tx2783和RTx436构建了高质量的染色体水平基因组组装。通过先进的单分子技术、长读长测序和光学图谱，与现有的Btx623 v3.0参考基因组相比，我们将平均序列连续性提高了19倍和11倍，分别为Tx2783和RTx436获得了19个和18个支架（N50分别为25.6和14.4）。我们的基因注释工作为Tx2783基因组产生了29612个蛋白质编码基因，为RTx436基因组产生了29265个蛋白质编码基因。与26个植物基因组（包括18个高粱基因组和8个外类群物种）的比较分析确定了约31210个蛋白质编码基因家族，其中约13956个是高粱特有的。利用18个高粱基因组中基因树的代表性模型，共鉴定出72579个泛基因，其中14%为核心基因，60%为软核心基因，26%为外壳基因。通过GO富集Pearson卡方检验，我们在Tx2783中鉴定出99个基因，在RTx436中鉴定出107个基因，这些基因在结合和代谢过程中表现出功能富集。在BTx623 Bionano图谱与BTx623 v3.1参考序列的比较中，我们检测到36个潜在的大倒位。令人惊讶的是，在将Tx2783或RTx436与BTx623 Bionano图谱进行比较时，这些倒位明显不存在。这些倒位大多位于着丝粒周围区域，该区域已知具有低复杂性区域且难以组装，这表明公共BTx623参考组装中存在潜在的人工产物。此外，以BTx623作为参考基因组时，与Tx2783相比，RTx436表现出另外324883个单核苷酸多态性（SNP）和16506个更多的插入/缺失（INDEL）。我们还对这两个基因组中的约348个核苷酸结合富含亮氨酸重复序列（NLR）抗病基因进行了表征。这些高质量的基因组为发现农艺性状和结构变异研究提供了宝贵的资源。
