Generation of reference genomes for Trypanosoma cruzi using PacBio sequencing

 描述（由申请人提供）：克氏锥虫基因组研究的最持久和最重要的障碍是缺乏可接受的参考基因组。有许多因素阻碍了产生高质量、完全组装的T。cruzi参考序列，包括T. cruzi基因组中的极端全基因组杂合性，选择用于测序和参考基因组组装的参考菌株（CL-Brener）中的极端全基因组杂合性，以及参考菌株中的遗传嵌合性。因此，T. Cruzi具有数百个缺口和装配崩溃区域，从未被完全装配，事实上，由于参考菌株的杂合性，Cruzi呈现为两个基因组-每条染色体呈现为两个版本，这两个版本在基因内容、两条染色体序列对齐的序列同一性、甚至成对的两条染色体的长度方面都有很大不同。这些问题极大地削弱了以前和正在进行的全基因组测序研究在这种人类疾病的病原体中的功效和力量，并且使得几乎任何涉及需要仔细检查参考基因组的努力都变得困难并且经常无效。我们建议用目前的T.首先，我们提出使用单分子、长读段（高达10 kb）测序来闭合缺口并扩展组装崩溃区域，所述组装崩溃区域在主要使用桑格读段（700 bp）组装原始CL-Brener参考基因组时发生。第二，我们选择T。用于测序的纯合且不具有嵌合基因组的Cruzi分离物以简化组装。第三，因为所有的T。cruzi品系来源于两个祖先的、不同的、纯合的、非嵌合的谱系，我们将产生每个"亲本"谱系的参考序列，以覆盖所有T.克鲁兹分离物。

 DESCRIPTION (provided by applicant): The most persistent and significant hindrance to genomic studies of Trypanosoma cruzi is the lack of an acceptable reference genome. There are a number of factors that have impeded progress in generating a high quality, fully assembled T. cruzi reference sequence, including high repeat content (50%) in the T. cruzi genome, extreme genome-wide heterozygosity in the reference strain (CL-Brener) that was chosen for sequencing and reference genome assembly, and genetic mosaicism in the reference strain. Thus, the current reference genome for T. cruzi has hundreds of gaps and regions of assembly collapse, has never been fully assembled and is, in fact, presented as two genomes due to the heterozygosity of the reference strain -each chromosome is presented as two versions differing substantially in terms of gene content, sequence identity where the two chromosome sequences align, and even lengths of the two chromosomes in a pair. These issues have greatly muted the efficacy and power of previous and ongoing whole genome sequencing studies in this causative agent of human disease, and have made virtually any endeavor involving the need for scrutinizing the reference genome difficult and frequently unproductive. We propose to tackle the issues with the current T. cruzi reference genome using three modifications of the approach originally used to generate it. First, we propose to use single-molecule, long-read (up to 10 kb) sequencing to close gaps and expand regions of assembly collapse that occurred when primarily Sanger reads (700 bp) were used to assemble the original CL-Brener reference genome. Second, we will select T. cruzi isolates for sequencing that are homozygous and do not have mosaic genomes to simplify the assemblies. Third, because all T. cruzi lines are derived from two ancestral, divergent, homozygous, non-mosaic lineages, we will generate reference sequences to each "parental" lineage in order to cover the scope of genetic content in all T. cruzi isolates.