A Widespread, Comparative Assessment of LTR-Retrotransposons - the Major Components of Plant Genomes

By far the major part of plant genomes consists of LTR-retrotransposons and their deteriorated remnants. They can cover over 80% of large genomes, like barley (5 Gbp) or wheat (17 Gbp), adding up to several Gbp of highly repetitive transposon derived sequences. These fast evolving 'junk' sequences define chromosomal architectures, they create phenotypic variations, are involved in epigenetic regulations and can be used to trace back genome evolution. Despite their importance LTR-retrotransposons are not consistently annotated across different plant species. But they are - due to their well defined structural properties - well amenable to bioinformatic de novo detection approaches.

Aimed at efficiently generating homogenous data sets suited for species wide comparisons, we established a streamlined LTR detection and analysis pipeline. The work flow starts with candidate detection via LTRharvest (genometools.org), followed by characterization modules - Pfam domain order for classification, insertion age, family clusters - and finally filter steps to remove non canonical elements. The resulting classified full length LTR-retrotransposon elements are then added to our plant repeat database PGSB REdat (pgsb.helmholtz-muenchen.de/plant/recat/http://pgsb.helmholtz-muenchen.de/plant/recat/) and used as templates to detect the multitude of deteriorated fragments within the genome assemblies.

After a brief introduction to the pipeline principles the talk will present results for a widespread selection of different plant species in respect to basic numbers, insertion age distributions, family sizes and broad to fine scaled chromosomal locations.