Recurrence of Chromosome Rearrangements and Reuse of DNA Breakpoints in the Evolution of the Triticeae Genomes

Chromosomal rearrangements (CRs) play important roles in karyotype diversity and speciation. While many CR breakpoints have been characterized at the sequence level in yeast, insects, and primates, little is known about the structure of evolutionary CR breakpoints in plant genomes, which are much more dynamic in genome size and sequence organization. Here, we report identification of breakpoints of a translocation between chromosome arms 4L and 5L of Triticeae, which is fixed in several species, including diploid wheat and rye, by comparative mapping and analysis of the draft genome and chromosome survey sequences of the Triticeae species. The wheat translocation joined the ends of breakpoints downstream of a WD40 gene on 4AL and a gene of the PMEI family on 5AL. A basic helix-loop-helix transcription factor gene in 5AL junction was significantly restructured. Rye and wheat share the same position for the 4L breakpoint, but the 5L breakpoint positions are not identical, although very close in these two species, indicating the recurrence of 4L/5L translocations in the Triticeae. Although barley does not carry the translocation, collinearity across the breakpoints was violated by putative inversions and/or transpositions. Alignment with model grass genomes indicated that the translocation breakpoints coincided with ancient inversion junctions in the Triticeae ancestor. Our results show that the 4L/5L translocation breakpoints represent two CR hotspots reused during Triticeae evolution, and support breakpoint reuse as a widespread mechanism in all eukaryotes. The mechanisms of the recurrent translocation and its role in Triticeae evolution are also discussed.

Reconstruction of ancestral genomes in presence of gene gain and loss.

Since most dramatic genomic changes are caused by genome rearrangements as well as gene duplications and gain/loss events, it becomes crucial to understand their mechanisms and reconstruct ancestral genomes of the given genomes. This problem was shown to be NP-complete even in the "simplest" case of three genomes, thus calling for heuristic rather than exact algorithmic solutions. At the same time, a larger number of input genomes may actually simplify the problem in practice as it was earlier illustrated with MGRA, a state-of-the-art software tool for reconstruction of ancestral genomes of multiple genomes. One of the key obstacles for MGRA and other similar tools is presence of breakpoint reuses when the same breakpoint region is broken by several different genome rearrangements in the course of evolution. Furthermore, such tools are often limited to genomes composed of the same genes with each gene present in a single copy in every genome. This limitation makes these tools inapplicable for many biological datasets and degrades the resolution of ancestral reconstructions in diverse datasets. We address these deficiencies by extending the MGRA algorithm to genomes with unequal gene contents. The developed next-generation tool MGRA2 can handle gene gain/loss events and shares the ability of MGRA to reconstruct ancestral genomes uniquely in the case of limited breakpoint reuse. Furthermore, MGRA2 employs a number of novel heuristics to cope with higher breakpoint reuse and process datasets inaccessible for MGRA. In practical experiments, MGRA2 shows superior performance for simulated and real genomes as compared to other ancestral genomes reconstruction tools. The MGRA2 tool is distributed as an open-source software and can be downloaded from GitHub repository http://github.com/ablab/mgra/. It is also available in the form of a web-server at http://mgra.cblab.org, which makes it readily accessible for inexperienced users.

Fast computation of a string duplication history under no-breakpoint-reuse

In this paper, we provide an O ( n log 2 n log log n log* n ) algorithm to compute a duplication history of a string under no-breakpoint-reuse condition. The motivation of this problem stems from computational biology, in particular, from analysis of complex gene clusters. The problem is also related to computing edit distance with block operations, but, in our scenario, the start of the history is not fixed, but chosen to minimize the distance measure.