xianran.li@usda.gov

Environmental conditions play a significant role in shaping phenotypic variation observed in natural conditions, but it remains challenging to parse out and integrate environmental contributions into genetic analyses and breeding programs. Two primary obstacles are acquiring data from a large number of environments and implementing intuitive analytics.

Variety testing trial has been an essential component for modern breeding programs over decades, these publicly accessible performance records, coupling weather database, could be repurposed to overcome the data acquisition hurdle.

Newly developed machine learning and deep learning algorithms, inspired by well-established genetics approaches, can be seamlessly integrated into the classic Joint regression model pioneered by Australian Scientists Finlay and Wilkinson over 60 years ago.

Combining these innovative data acquisition and analytics strategies empowers exploit weather conditions to wheat performance.