The honey bee reference genome, HAv3.1, was produced from a commercial line sample that was thought to have a largely dominant Apis mellifera ligustica genetic background. Apis mellifera mellifera, often referred to as the black bee, has a separate evolutionary history and is the original type in western and northern Europe. Growing interest in this subspecies for conservation and non-professional apicultural practices, together with the necessity of deciphering genome backgrounds in hybrids, triggered the necessity for a specific genome assembly. Moreover, having several high-quality genomes is becoming key for taking structural variations into account in pangenome analyses. Pacific Bioscience technology long reads were produced from a single haploid black bee drone. Scaffolding contigs into chromosomes was done using a high-density genetic map. This allowed for re-estimation of the recombination rate, which was over-estimated in some previous studies due to mis-assemblies, which resulted in spurious inversions in the older reference genomes. The sequence continuity obtained was very high and the only limit towards continuous chromosome-wide sequences seemed to be due to tandem repeat arrays that were usually longer than 10 kb and that belonged to two main families, the 371 and 91 bp repeats, causing problems in the assembly process due to high internal sequence similarity. Our assembly was used together with the reference genome to genotype two structural variants by a pangenome graph approach with Graphtyper2. Genotypes obtained were either correct or missing, when compared to an approach based on sequencing depth analysis, and genotyping rates were 89 and 76% for the two variants. Our new assembly for the Apis mellifera mellifera honey bee subspecies demonstrates the utility of multiple high-quality genomes for the genotyping of structural variants, with a test case on two insertions and deletions. It will therefore be an invaluable resource for future studies, for instance by including structural variants in GWAS. Having used a single haploid drone for sequencing allowed a refined analysis of very large tandem repeat arrays, raising the question of their function in the genome. High quality genome assemblies for multiple subspecies such as presented here, are crucial for emerging projects using pangenomes.

中文翻译：

---

黑蜜蜂基因组：对特定结构元素的见解和迈向全基因组的第一步


蜜蜂参考基因组 HAv3.1 是从商业品系样本中产生的，该样本被认为具有很大程度上显性的意大利蜜蜂遗传背景。意大利蜜蜂（Apis mellifera mellifera），通常被称为黑蜂，有独立的进化历史，是西欧和北欧的原始类型。人们对这个亚种的保护和非专业养蜂实践的兴趣日益浓厚，加上破译杂交品种基因组背景的必要性，引发了对特定基因组组装的必要性。此外，拥有几个高质量的基因组正成为在泛基因组分析中考虑结构变异的关键。 Pacific Bioscience 技术的长读段是由单个单倍体黑蜂雄蜂产生的。使用高密度遗传图谱将重叠群支架固定到染色体中。这允许重新估计重组率，在之前的一些研究中，由于组装错误，重组率被高估，导致旧参考基因组中出现虚假倒位。获得的序列连续性非常高，并且对连续染色体范围序列的唯一限制似乎是由于串联重复序列阵列通常长于 10 kb 并且属于两个主要家族，即 371 bp 和 91 bp 重复序列，从而导致了以下问题：由于内部序列高度相似，组装过程变得更加容易。我们的组装与参考基因组一起使用，通过 Graphtyper2 的泛基因组图方法对两种结构变异进行基因分型。与基于测序深度分析的方法相比，获得的基因型要么正确，要么缺失，两种变体的基因分型率分别为 89% 和 76%。 我们对 Apis mellifera mellifera 蜜蜂亚种的新组装证明了多个高质量基因组在结构变异基因分型中的实用性，并提供了两个插入和删除的测试案例。因此，它将成为未来研究的宝贵资源，例如将结构变异纳入 GWAS 中。使用单个单倍体雄蜂进行测序可以对非常大的串联重复阵列进行精细分析，从而提出了它们在基因组中的功能问题。对于多个亚种的高质量基因组组装（如此处所示）对于使用泛基因组的新兴项目至关重要。