W457
Strategies For Sequence Improvement Of BAC-Based Wheat Chromosome Arm Assemblies

Date: Tuesday, January 15, 2013
Time: 2:40 PM
Room: Golden Ballroom
Jane Rogers , International Wheat Genome Sequencing Consortium, Cambridge, United Kingdom
Deepali Vasoya , The Genome Analysis Centre, Norwich, England
Darren Heavens , The Genome Analysis Centre, Norwich, England
Adriana Alberti , CEA - Genoscope, Evry, France
Arnaud Couloux , CEA - Genoscope, Evry, France
Valerie Barbe , CEA - Genoscope, Evry, France
Frederic Choulet , INRA GDEC, Clermont-Ferrand, France
Sébastien Theil , INRA GDEC, Clermont-Ferrand, France
Catherine Feuillet , INRA GDEC, Clermont-Ferrand, France
Mario Caccamo , The Genome Analysis Centre, Norwich, United Kingdom
Matt Clark , The Genome Analysis Centre, Norwich, United Kingdom
To overcome the challenges that the large size (17Gb), hexaploidy and the large repeat content (est. >85%) pose for assembling a high utility reference sequence of bread wheat (Triticum aestivum), The International Wheat Genome Sequencing Consortium has adopted a ‘divide and conquer’ approach. Physical maps are first built from Bacterial Artificial Chromosomes (BACs) produced from the DNA of individual flow sorted chromosome arms and the minimal tiling path (MTP) of clones forms the substrate for sequence generation. Although this reduces the sequence assembly problem to 100-150kb, the requirement to use next generation sequencing in highly parallelised mode to make the sequencing affordable, means that the initial product of BAC sequencing and assembly is still a large number of small contigs that are difficult to position in their genomic context. We have been looking at approaches to improve chromosome assemblies, using BACs  from the MTP of chromosome 3B, which has been sequenced recently using a hybrid 454/Illumina strategy. We have assembled two different BAC pools containing genes, pseudogenes and transposable elements in combinations that are representative of different regions in the wheat genome. New tools e.g. SMRTanalysis, Celera Assembler v7.0 and ALLPATHS-LG (revision 41240 or higher) have been used to combine 454, Illumina (paired end and mate-pair) sequences with Pacific Biosciences long read sequence data. Contig N50 values and length of accurate contiguous sequences will be reported for each of the pools using different combinations of data and assembly tools. Combining different sequencing technologies and assembly tools can improve assemblies and may assist in sequencing the wheat genome and other highly complex genomes.