BACKGROUND Currently available short read genome assemblies of the tetraploid protozoan parasite Giardia intestinalis are highly fragmented, highlighting the need for improved genome assemblies at a reasonable cost. Long nanopore reads are well suited to resolve repetitive genomic regions resulting in better quality assemblies of eukaryotic genomes. Subsequent addition of highly accurate short reads to long-read assemblies further improves assembly quality. Using this hybrid approach, we assembled genomes for three Giardia isolates, two with published assemblies and one novel, to evaluate the improvement in genome quality gained from long reads. We then used the long reads to predict structural variants to examine this previously unexplored source of genetic variation in Giardia. METHODS With MinION reads for each isolate, we assembled genomes using several assemblers specializing in long reads. Assembly metrics, gene finding, and whole genome alignments to the reference genomes enabled direct comparison to evaluate the performance of the nanopore reads. Further improvements from adding Illumina reads to the long-read assemblies were evaluated using gene finding. Structural variants were predicted from alignments of the long reads to the best hybrid genome for each isolate and enrichment of key genes was analyzed using random genome sampling and calculation of percentiles to find thresholds of significance. RESULTS Our hybrid assembly method generated reference quality genomes for each isolate. Consistent with previous findings based on SNPs, examination of heterozygosity using the structural variants found that Giardia BGS was considerably more heterozygous than the other isolates that are from Assemblage A. Further, each isolate was shown to contain structural variant regions enriched for variant-specific surface proteins, a key class of virulence factor in Giardia. CONCLUSIONS The ability to generate reference quality genomes from a single MinION run and a multiplexed MiSeq run enables future large-scale comparative genomic studies within the genus Giardia. Further, prediction of structural variants from long reads allows for more in-depth analyses of major sources of genetic variation within and between Giardia isolates that could have effects on both pathogenicity and host range.

中文翻译：

---

鉴定贾第鞭毛虫的杂种组合，并预测广泛的内部隔离结构变异。

背景技术四倍体原生动物寄生虫贾第鞭毛虫肠的当前可用的短读基因组组装体高度碎片化，强调了以合理的成本改进基因组组装体的需要。长的纳米孔读段非常适合解析重复的基因组区域，从而产生更好质量的真核基因组。随后在长时间读取的程序集中添加了高度精确的短读取，进一步提高了程序集质量。使用这种杂交方法，我们组装了三个贾第虫分离株的基因组，其中两个具有公开的程序集和一种新颖的化合物，以评估从长时间阅读中获得的基因组质量的提高。然后，我们使用长读物预测结构变异，以检查贾第鞭毛虫的这一以前未曾探索过的遗传变异来源。用MinION读取每个分离株的方法 我们使用多个专门进行长阅读的汇编程序组装了基因组。装配指标，基因发现以及与参考基因组的整个基因组比对，使得可以直接比较以评估纳米孔读数的性能。使用基因发现评估了将Illumina读段添加到长读组件中的进一步改进。通过对每个分离株的长读物与最佳杂交基因组的比对，预测结构变异，并使用随机基因组采样和百分位数计算来分析关键基因的富集，以找到重要的阈值。结果我们的混合装配方法为每个分离株产生了参考质量的基因组。与先前基于SNP的发现一致，使用结构变体检查杂合性后发现，贾第鞭毛虫BGS比来自组合A的其他分离株的杂合性要高得多。此外，每个分离株均显示出富含变异体特异性表面蛋白（一种重要的毒力因子）的结构变异区在贾第虫。结论从单个MinION运行和多重MiSeq运行生成参考质量基因组的能力，使将来在贾第鞭毛虫属中进行大规模的比较基因组研究。此外，通过长读对结构变异的预测可以对贾第鞭毛虫分离株内部和之间的遗传变异的主要来源进行更深入的分析，这可能会对致病性和寄主范围产生影响。每个分离物都显示出富含变异体特异性表面蛋白（贾第鞭毛虫中的关键毒力因子）的结构变异区。结论从单个MinION运行和多重MiSeq运行生成参考质量基因组的能力，使将来在贾第鞭毛虫属中进行大规模的比较基因组研究。此外，通过长读对结构变异的预测可以对贾第鞭毛虫分离株内部和之间的遗传变异的主要来源进行更深入的分析，这可能会对致病性和寄主范围产生影响。每个分离物都显示出富含变异体特异性表面蛋白（贾第鞭毛虫中的关键毒力因子）的结构变异区。结论从单个MinION运行和多重MiSeq运行生成参考质量基因组的能力，使将来在贾第鞭毛虫属中进行大规模的比较基因组研究。此外，通过长读对结构变异的预测可以对贾第鞭毛虫分离株内部和之间的遗传变异的主要来源进行更深入的分析，这可能会对致病性和寄主范围产生影响。结论从单个MinION运行和多重MiSeq运行生成参考质量基因组的能力，使将来在贾第鞭毛虫属中进行大规模的比较基因组研究。此外，通过长读对结构变异的预测可以对贾第鞭毛虫分离株内部和之间的遗传变异的主要来源进行更深入的分析，这可能会对致病性和寄主范围产生影响。结论从单个MinION运行和多重MiSeq运行生成参考质量基因组的能力，使将来在贾第鞭毛虫属中进行大规模的比较基因组研究。此外，通过长读对结构变异的预测可以对贾第鞭毛虫分离株内部和之间的遗传变异的主要来源进行更深入的分析，这可能会对致病性和寄主范围产生影响。