Virtual StatsPD@Waite meeting

Every month, the professional development meetings of statisticians and data scientists at Waite, known as StatsPD@Waite, bring together specialists in various aspects of data sciences in agriculture from Waite, Roseworthy and Adelaide.

Please join us for the next StatsPD@Waite seminar where Dr Eric G. Dinglasan, Senior Research Fellow, QAAFI, The University of Queensland will present on “FastStack: evolutionary computing-guided breeding to develop future crops”.

Also please note that the StatsPD@Waite meetings are recorded. If you have a question to the speaker but had rather not be recorded, please send your question via chat during the meeting and it will be asked on your behalf.

Please email Beata Sznajder with questions or for details of the Zoom meeting.

Title: FastStack: evolutionary computing-guided breeding to develop future crops

Presenter:Dr Eric G. Dinglasan, Senior Research Fellow, QAAFI, The University of Queensland

A significant challenge for elite plant breeding germplasm is to increase rates of genetic gain without exhausting the genetic diversity. Selecting parents to be used in designing crosses is a critical step in the breeding cycle that determines the success of a breeding program. A standard practice is to select superior lines based on the overall ‘merit’ and use as parents to develop new populations. This leads to rapid gain, however, also unintentionally accelerates loss of genetic variation. Moreover, ‘low merit’ lines are discarded and often carry ‘desirable’ alleles of high value, making these desirable alleles underutilised. This is particularly challenging as loss of genetic variability can limit long-term gains for new traits that are becoming important as climate changes. Utilisation of pre-breeding germplasm and introgression of few traits is an alternative approach to introduce new diversity. However, it is a slow and challenging process.

Here, we investigate genetic algorithm and other types of evolutionary computing algorithm (i.e. a type of AI) approach to rapidly stack chromosome segments to build ultimate genotypes. The aim is to select parental crosses which “stack” the most desirable chromosome segments (i.e. haplotype with the most favourable effect on the trait) in the population at each genome location into the ultimate individual, with the highest possible genetic merit for the trait.

I will present 2 case studies: 1. Using a commercial wheat breeding data to develop a proof-of-concept of chromosome segment stacking approach to help increase long-term genetic gain while simultaneously maintaining genetic diversity in the breeding population; and 2. Identifying exotic and resistance haplotypes to help diversify net blotch resistance for barley genetic resistance improvement in Australia.