Segmental duplications: organization and impact within the current human genome project assembly

Segmental duplications play fundamental roles in both genomic disease and gene evolution. To understand their organization within the human genome, we have developed the computational tools and methods necessary to detect identity between long stretches of genomic sequence despite the presence of high copy repeats and large insertion-deletions. Here we present our analysis of the most recent genome assembly (January 2001) in which we focus on the global organization of these segments and the role they play in the whole-genome assembly process. Initially, we considered only large recent duplication events that fell well-below levels of draft sequencing error (alignments 90%-98% similar and > or =1 kb in length). Duplications (90%-98%; > or =1 kb) comprise 3.6% of all human sequence. These duplications show clustering and up to 10-fold enrichment within pericentromeric and subtelomeric regions. In terms of assembly, duplicated sequences were found to be over-represented in unordered and unassigned contigs indicating that duplicated sequences are difficult to assign to their proper position. To assess coverage of these regions within the genome, we selected BACs containing interchromosomal duplications and characterized their duplication pattern by FISH. Only 47% (106/224) of chromosomes positive by FISH had a corresponding chromosomal position by comparison. We present data that indicate that this is attributable to misassembly, misassignment, and/or decreased sequencing coverage within duplicated regions. Surprisingly, if we consider putative duplications >98% identity, we identify 10.6% (286 Mb) of the current assembly as paralogous. The majority of these alignments, we believe, represent unmerged overlaps within unique regions. Taken together the above data indicate that segmental duplications represent a significant impediment to accurate human genome assembly, requiring the development of specialized techniques to finish these exceptional regions of the genome. The identification and characterization of these highly duplicated regions represents an important step in the complete sequencing of a human reference genome.     

Role of genomic architecture in PLP1 duplication causing Pelizaeus-Merzbacher disease

Genomic architecture, higher order structural features of the human genome, can provide molecular substrates for recurrent sub-microscopic chromosomal rearrangements, or may result in genomic instability by forming structures susceptible to DNA double-strand breaks. Pelizaeus-Merzbacher disease (PMD) is a genomic disorder most commonly arising from genomic duplications of the dosage-sensitive proteolipid protein gene (PLP1). Unlike many other genomic disorders that result from non-allelic homologous recombination utilizing flanking low-copy repeats (LCRs) as substrates, generating a common and recurrent rearrangement, the breakpoints of PLP1 duplications have been reported not to cluster, yielding duplicated genomic segments of varying lengths. This suggests a distinct molecular mechanism underlying PLP1 duplication events. To determine whether structural features of the genome also facilitate PLP1 duplication events, we analyzed extensively the genomic architecture of the PLP1 region and defined several novel LCRs (LCR-PMDs). Array comparative genomic hybridization showed that PLP1 duplication sizes differed, but revealed a subgroup of patients with apparently similar PLP1 duplication breakpoints. Pulsed-field gel electrophoresis analysis using probes adjacent to the LCR-PMDs detected unique recombination-specific junction fragments in 12 patients, enabled us to associate the LCR-PMDs with breakpoint regions, and revealed rearrangements inconsistent with simple tandem duplications in four patients. Two-color fluorescence in situ hybridization was consistent with directly oriented duplications. Our study provides evidence that PLP1 duplication events may be stimulated by LCRs, possibly non-homologous pairs at both the proximal and distal breakpoints in some cases, and further supports an alternative role of genomic architecture in rearrangements responsible for genomic disorders.     

Gene duplication and the structure of eukaryotic genomes

A simple method for understanding how gene duplication has contributed to genomic structure was applied to the complete genomes of Caenorhabditis elegans, Drosophila melanogaster, and yeast Saccharomyces cerevisiae. By this method, the genes belonging to gene families (the paranome) were identified, and the extent of sharing of two or more families between genomic windows was compared with that expected under a null model. The results showed significant evidence of duplication of genomic blocks in both C. elegans and yeast. In C. elegans, the five block duplications identified all occurred intra-chromosomally, and all but one occurred quite recently. In yeast, by contrast, 39 duplicated blocks were identified, and all but one of these was inter-chromosomal. Of these 39 blocks, 28 showed evidence of ancient duplication, possibly as a result of an ancient polyploidization event. By contrast, three blocks showed evidence of very recent duplication, while seven others showed a mixture of ancient and recent duplication events. Thus, duplication of genomic blocks has been an ongoing feature of yeast evolution over the past 200--300 million years.     

Genomic organization of the sex-determining and adjacent regions of the sex chromosomes of medaka

Sequencing of the human Y chromosome has uncovered the peculiarities of the genomic organization of a heterogametic sex chromosome of old evolutionary age, and has led to many insights into the evolutionary changes that occurred during its long history. We have studied the genomic organization of the medaka fish Y chromosome, which is one of the youngest heterogametic sex chromosomes on which molecular data are available. The Y specific and adjacent regions were sequenced and compared to the X. The male sex-determining gene, dmrt1bY, appears to be the only functional gene in the Y-specific region. The Y-specific region itself is derived from the duplication of a 43-kb fragment from linkage group 9. All other coduplicated genes except dmrt1bY degenerated. The Y-specific region has accumulated large stretches of repetitive sequences and duplicated pieces of DNA from elsewhere in the genome, thereby growing to 258 kb. Interestingly the non-recombining part of the Y did not spread out considerably from the original duplicated fragment, possibly because of a large sequence duplication bordering the Y-specific fragment. This may have conserved the more ancestral structure of the medaka Y and provides insights into some of the initial processes of Y chromosome evolution.     

Dissecting the Structure and Mechanism of a Complex Duplication–Triplication Rearrangement in the DMD Gene

Pathogenic complex genomic rearrangements are being increasingly characterized at the nucleotide level, providing unprecedented opportunities to evaluate the complexities of mutational mechanisms. Here, we report the molecular characterization of a complex duplication–triplication rearrangement involving exons 45–60 of the DMD gene. Inverted repeats facilitated this complex rearrangement, which shares common genomic organization with the recently described duplication‐inverted triplication–duplication (DUP–TRP/INV‐DUP) events; specifically, a 690‐kb region comprising DMD exons from 45 to 60 was duplicated in tandem, and another 46‐kb segment containing exon 51 was inserted inversely in between them. Taking into consideration (1) the presence of a predicted PRDM9 binding site in the near vicinity of the junction involving two inverted L1 elements and (2) the inherent properties of X–Y chromosome recombination during male meiosis, we proposed an alternative two‐step model for the generation of this X‐linked DMD DUP–TRP/INV‐DUP event.     

Gene conversion,unequal crossing-over and mispairing at a non-tandem duplication during meiosis of Saccharomyces cerevisiae

Summary We have developed a novel system to examine conversion, exchange and mispairing involving a nontandem duplication of the ade8 locus in yeast by monitoring the segregation of heterozygous markers between the duplicated sequence. Plasmid Yrp 17 carries the yeast selectable markers URA3 ⁺ and TRP1 ⁺. Yrpl7 derivatives with a 4 kb insert carrying ade8-18 were used to clone the mutations trpl-1 and ura3-1 by gap repair. Integrants of the resulting plasmids at the Ade8 locus were crossed to yield diploid hybrids with a non-tandem duplication of Ade8 and heterozygosity for the plasmid markers between the duplicated sequences. 1192 complete, unselected asci were analyzed and 270 exhibiting recombination of the markers contributed by the plasmid were analyzed by Southern transfers to detect changes in plasmid sequences. Twenty-seven tetrads had unequal homologous exchanges and five had unequal sister-chromatid exchanges. Seven tetrads carry an additional copy of the integrated plasmid and ten are missing one. We propose that these two classes represent conversions of the entire 11 kb plasmid, which occur after misalignment and formation of an unpaired loop. Mispairing is a frequent event, and occurs in approximately fifty percent of all meioses. The system described provides a means to determine the meiotic rules of conversion, exchange and pairing for duplicated DNA sequences.     

Genotype-phenotype correlation in five Pelizaeus-Merzbacher disease patients with PLP1 gene duplications

Pelizaeus–Merzbacher disease (PMD) is an X-linked myelination disorder most frequently caused by duplication of a genomic segment of variable length containing the PLP1 gene. We studied five PMD male patients affected by the classic PMD form carrying a PLP1 gene duplication. On the basis of clinical and neuroradiological features, two of the five patients appeared to be the most severely affected. In order to establish a possible genotype–phenotype correlation, the extent of the duplication was determined in each patient and in the respective mother by quantifying the copy number of genomic markers surrounding the PLP1 gene by a real-time PCR-based approach. Duplications, ranging in size from 167–195 to 580–700 kb, were in the same genomic interval of the majority of the reported duplications. The extent of the duplicated genomic segments does not correlate with the clinical severity. Interestingly enough, each duplication had one of the two breakpoints in or near to low copy repeats (LCRs), supporting recent evidence concerning a possible role of LCRs in the generation of the duplications in PMD.     

Retrotransposon Ty1 integration targets specifically positioned asymmetric nucleosomal DNA segments in tRNA hotspots

The Saccharomyces cerevisiae genome contains about 35 copies of dispersed retrotransposons called Ty1 elements. Ty1 elements target regions upstream of tRNA genes and other Pol III-transcribed genes when retrotransposing to new sites. We used deep sequencing of Ty1-flanking sequence amplicons to characterize Ty1 integration. Surprisingly, some insertions were found in mitochondrial DNA sequences, presumably reflecting insertion into mitochondrial DNA segments that had migrated to the nucleus. The overwhelming majority of insertions were associated with the 5' regions of Pol III transcribed genes; alignment of Ty1 insertion sites revealed a strong sequence motif centered on but extending beyond the target site duplication. A strong sequence-independent preference for nucleosomal integration sites was observed, in distinction to the preferences of the Hermes DNA transposon engineered to jump in yeast and the Tf1 retrotransposon of Schizosaccharomyces pombe, both of which prefer nucleosome free regions. Remarkably, an exquisitely specific relationship between Ty1 integration and nucleosomal position was revealed by alignment of hotspot Ty1 insertion position regions to peak nucleosome positions, geographically implicating nucleosomal DNA segments at specific positions on the nucleosome lateral surface as targets, near the "bottom" of the nucleosome. The specificity is observed in the three tRNA 5'-proximal nucleosomes, with insertion frequency dropping off sharply 5' of the tRNA gene. The sites are disposed asymmetrically on the nucleosome relative to its dyad axis, ruling out several simple molecular models for Ty1 targeting, and instead suggesting association with a dynamic or directional process such as nucleosome remodeling associated with these regions.     

Identical rearrangement of NSP3 genes found in three independently isolated virus clones derived from mixed infection and multiple passages of Rotaviruses

Three rotavirus variants with a rearranged RNA segment derived from the NSP3 gene were isolated in three independent experiments of coinfection and multiple passages of simian rotavirus strain SA11 and single-VP7-gene- or NSP1-gene-substitution reassortants having genetic background of SA11. Sequence analysis indicated that the three rearranged NSP3 genes had almost identical sequences and genomic structures organized by partial duplication of the open reading frame in a head-to-tail orientation following the termination codon. The junction site of the original NSP3 gene (first copy) and the duplicated portion (second copy) was identical among the three rearranged genes, while a direct repeat, i.e., a homologous sequence between the first copy and second template for duplication, typically located at the junction site, was not detected. However, short similar sequences were present at the end of the first copy and beginning of the second copy. These findings suggest that rearrangement of the NSP3 gene may occur at a certain preferential site which is related to sequence similarity between 3′-untranslated region and a region near the 5′-end of ORF.