Application of AI for wheat yield improvement



External Member

Khandaker Asif Ahmed, CSIRO (


In response to the challenges facing 21st century agriculture, advanced science and technology approaches need to be developed and adopted to provide improved avenues for growth and food security. Machine learning will be part of this solution. Wheat is a major source of human nutrition. This project aims to develop and test a deep-learning-based AI model for predicting wheat yield component traits from multiple ‘omic data, including genome (Single Nucleotide polymorphism -SNPs) and transcriptome, and identify causal biomolecules leading to high-yield. The resulting model will demonstrate feasibility of applying ML&AI to biological data toward the improvement of crop yield by enabling useful tasks such as predicting varietal productivity in different environments.


  1. Experience in computer science or knowledge of Python coding language is essential.
  2. Basic understanding of biological data (e.g. genomic, transcriptomics) is an advantage.

Background Literature

  1. Sandhu et al. (2021). Deep Learning for Predicting Complex Traits in Spring Wheat Breeding ProgramPlant Systems and Synthetic Biology, 11(613325):2084.
  2. Wang et al. (2017). Transcriptome Association Identifies Regulators of Wheat Spike ArchitecturePlant Physiology, 175(2):746-757.

Updated:  10 August 2021/Responsible Officer:  Dean, CECS/Page Contact:  CECS Marketing