How Next-Generation Sequencing Has Aided Our Understanding of the Sequence Composition and Origin of B Chromosomes

Abstract

Accessory, supernumerary, or-most simply-B chromosomes, are found in many eukaryotic karyotypes. These small chromosomes do not follow the usual pattern of segregation, but rather are transmitted in a higher than expected frequency. As increasingly being demonstrated by next-generation sequencing (NGS), their structure comprises fragments of standard (A) chromosomes, although in some plant species, their sequence also includes contributions from organellar genomes. Transcriptomic analyses of various animal and plant species have revealed that, contrary to what used to be the common belief, some of the B chromosome DNA is protein-encoding. This review summarizes the progress in understanding B chromosome biology enabled by the application of next-generation sequencing technology and state-of-the-art bioinformatics. In particular, a contrast is drawn between a direct sequencing approach and a strategy based on a comparative genomics as alternative routes that can be taken towards the identification of B chromosome sequences.

Keywords: B chromosome; evolution; next generation sequencing; supernumerary chromosome.

Figure 1

Direct and indirect methods used to identify B chromosome sequences using next generation sequencing (NGS). Strategy 1: the direct method. This approach requires a prior step, in which the B chromosomes are isolated either by micro-dissection or by flow-sorting. Strategy 2: the indirect method. This method requires the acquisition of sequence data from both an individual carrying a B chromosome(s) (+B dataset) and a related individual lacking any B chromosome(s) (0B dataset). The two datasets are compared using three alternative methods. In “similarity-based read clustering”, a graphically based analysis is performed using, for example, the RepeatExplorer pipeline. Sequence information is transformed into graphical structures (vertices correspond to sequence reads and edges characterize the overlap between reads). Differences (presence/absence of sequence reads) in the 0B and +B datasets affect the clusters, and are used to distinguish B chromosome sequences. The two-colored circles indicate reads containing sequences from 0B and +B probes. The “coverage ratio analysis” requires an initial alignment of reads, using an alignment pipeline such as Burrows-Wheeler Alignment tool (BWA). Differences in the read coverage ratio indicate B chromosome-derived candidate regions. The pink section illustrates an example of a putative candidate region, which features the absence of reads in the 0B dataset and their presence in the +B dataset. In the “k-mer frequency ratio analysis” approach, a program such as the Kmasker pipeline identifies differences in the k-mer frequency ratio. The illustration shows an example of a B chromosome segment (shown in pink) in which the k-mer frequency is low or zero in the 0B dataset, but high in the +B dataset. Both the coverage ratio and k-mer frequency ratio analyses, but not the similarity-based read clustering approach, require a reference sequence.