The in silico Chromosome Arm Assigner CarmA: A Component of the “Divide and Conquer” Approach to Tackle Large Plant Genomes

For the large and highly repetitive genomes of crop cereals like barley (5 Gb), rye (7 Gb) or the hexaploid wheat (17 Gb) a complete genome assembly still poses major challenges. Next generation sequences, derived from flow cytometric chromosome arm sorting, are a valuable new resource, that can be used to divide the complexity of the assembly strategies into smaller more manageable tasks. The bioinformatic tool CarmA (Chromosome arm Assigner) was developed to support those efforts with an efficient high throughput anchoring of unassigned sequence resources, like WGS assemblies or transcripts to specific chromosome arms. The approach exploits the sequence similarity between arm specific low copy regions of chromosome sorted reads and the respective query sequence. As proof of principle and to identify optimal parameter settings for different input scenarios CarmA was validated with simulated chromosome arm derived reads from assembled reference genomes. The positive predictive value (PPV) turned out to be consistently very high (>0.9)  throughout different input settings. The sensitivity varies between 0.4 and 0.8. Here a sufficient read coverage (> 1x) and longer queries (>500 bp) are crucial factors to obtain useful values.
CarmA has been successfully applied in the barley genome project [1] to guide and crosscheck the integration of diverse sequence resources with genetic markers and FPC data into a common chromosomal framework. Similar benefits can be expected for the IWGSC wheat genome project, which is also based on chromosome arm sorted sequences. For wheat the dissection of the highly similar A, B and D sub-genomes poses an additional layer of difficulties, that can be tackled with CarmA.

[1] The International Barley Genome Sequencing Consortium, A physical, genetic and functional sequence assembly of the barley genome, Nature 17.October 2012