Blog icon

Dr Shannon Dillon

Principal Research Scientist & Project Leader

Contact details:

GPO BOX 1700


Shannon is a biologist, geneticist and data scientist. Her skills and interests sit at the interface of quantitative genomics, GWAS, machine learning and population genetics. Her research focuses on the integration of multi-layered data streams to enhance our understanding of factors determining adaptive and productivity traits in plants, and the development of tools and approaches that exploit data to inform decision making in applied settings.

With an exceptional team of researchers at CSRIO, Shannon is applying multi ‘omic data platforms to uncover how genomic and environmental variation, the interaction of these factors and epistasis, control phenotypic traits in crop plants such as wheat and canola. In particular we have been exploring the drivers of flowering time in these systems, with a focus on delivering approaches that exploit biological data at the population scale to enhance crop productivity.

Shannon currently co-leads an MLAI Future Science Platform activity, Bioprediction, which is developing, extending and leveraging machine learning methods to transform how biological data and domain knowledge are utilised to enhance plant and animal breeding.

Other Interests

Research projects:

ML/AI Future Science Platform:
Transforming Biological Production Systems, 2019-2022 (activity area leader)
This activity area will develop, extend and leverage machine learning methods to solve biological problems by bringing together multiple information streams and data types, particularly phenotypic, genomic, other ‘omic and environmental data coupled with domain knowledge. The long-term goal of this activity is to deliver solutions that transform how data and knowledge are utilised in plant and animal breeding.

Optimising Canola Phenology for Australian Target Production Environments Grains:
Research and Development Corporation (CSP1901-002RTX), 2019-2022 (lead researcher)
Increased production and profitability of canola (Brassica napus) in Australia can be achieved by better matching phenology with the growing environment. We are bringing together genomic, other ‘omic and environmental data on a diverse set of modern Australian and ancestral canola varieties grown in key cropping regions, to resolve genetic and interaction effects underpinning phenological traits (e.g. flowering time) in canola. Improved understanding of these factors will enhance management and breeding of Australian canola.

Multi ‘omic Genome to phenome platform in Australian Wheat:
CSIRO Strategic Funding, 2015-2019 (lead researcher)
We are applying statistical methods including machine learning to integrate phenotypic, environmental, genomic and other high dimensional ‘omic (e.g. transcriptome) data layers at scale to resolve biological factors and interactions underpinning trait variation (e.g. flowering time). This project links directly with Australian wheat breeders (AGT) where we are working to optimise integration of ‘omic data streams into productivity assessments and breeding selection.

Visualising multilayered ‘omic’, phenotype and environmental data for biological inference:
eResearch projects, 2019 (lead researcher)
This project will establish an accessible tool set for visualisation of machine learning model outputs (e.g. associations, pathways, networks) to maximising the value we extract from our data and enable its timely and effective communication.

Machine Learning in Genome Biology – Cutting Edge Science Symposium:
CSIRO Research Office, 2018 – 2019 (symposium chair)
We hosted a three day symposium in Kiama, NSW, with key local and international speakers, from Tuesday the 9th to Thursday the 11th of April 2019. This workshop addressed advances and challenges in the application of “artificial intelligence” methods for feature extraction, assessment of interactions (GxE, GxG) and effective integration of ‘omic, environmental and trait/diagnostic data for phenotypic prediction.

National Phenology Initiative:
Grains Research and Development Corporation (ULA1806-004RTX - Hunt), 2018 – 2021
Crop performance models (e.g. APSIM) for Australian wheat are underpinned by a set of parameters derived from key growth and developmental traits in specific environments (e.g. phenology). The project will utilise genome wide SNP as well as known genetic markers for phenology QTL to examine how effectively parameters can be predicted from genomic information, and explore methods for integration into process based crop models.

Pedigree based association genetic analysis of wheat phenology:
Grains Research and Development Corporation (CSP00183 - Trevaskis), 2015 – 2018
We are applying machine learning methods to resolve gene, epistatic and gene by environment interactions underpinning flowering time variation in the Australian wheat pedigree, exploiting field experiments spanning thousands of varieties and trial sites collated over more than two decades.

Forests for the Future: making the most of a high CO2 world:
SIEF (Science Industry Endowment Fund) 2013 – 2018 (genomics lead)
Recent advances in eucalypt genome sequencing and plant phenomics provide pathways to screen responses to elevated CO2. Using Eucalyptus as a proof of concept we are assessing rapid screening techniques to predict the performance of eucalypt plantings under high CO2, to develop, test and deploy decision tools, including genomic based markers, for the forestry sector.

Evolutionary adaptation and adaptive plasticity in eucalypts:
CSIRO Transformational Biology Catalytic Project, 2012-2015
Rising CO2 will have consequences for the productivity and management of the world’s forests. Intraspecific variation in plasticity could impact responses to CO2 rise across species distributions, however remains poorly characterised in trees. Using a novel quantitative transcriptomics approach we explore genomic factors, including pre-existing adaptations, underlying plant physiological responses to rising CO2, in a continent wide sample of the river red gum.

Genome wide assessment of landscape level adaptation in eucalypts:
CSIRO Transformational Biology Catalytic Project, 2009 - 2013
We have assessed sample of genome wide SNP diversity in ~500 genotypes of the iconic riparian tree, the River Red Gum (Eucalyptus camaldulensis) spanning the length of the Murray River. We probed for signatures of adaptation to fine-scale environmental gradients within (and between) riparian sites, and link these patterns to variation in adaptive traits.

Community and Corporate Citizenship

  • 2015-2018

    Vacation scholar program – Agriculture and Food Comittee Chair. Our Undergraduate Vacation Scholarships are run over the Australian summer holidays and offers high achieving and promising undergraduate students the opportunity to collaborate with leading CSIRO scientists in our world class facilities.