With the ever-increasing ease of generating more and higher quality genomic data, annotation remains a fundamentally hard problem that stands between sequencing and insight. Most annotation is carried out using automated, homology-based, tools. These programs struggle to properly recognize and characterize genes that do not have closely related sequences in reference databases and many of them are explicitly coded to ignore repetitive annotations. This leads to unusually poor results from standard annotation pipelines for novel genes, those that evolve rapidly at the sequence level, and multicopy gene families.

Dr Ivan Koludarov is a Humboldt Research Fellow in the Technical University of Munich's Department of Scientific Computing and his research on venom genes lead him to confront these issues head on. His current research projects focus on the thorough and careful annotation of venom genes in high-quality genomes which enables him to examine the microsyntenic patterns of these gene families to reconstruct detailed evolutionary histories across narrow and broad taxonomic scales. Ivan will present a two-day workshop in Canberra on his genome annotation methods and how to apply them generally to multicopy gene families which are difficult to annotate (not just venoms).