Skip to main content

GWAS, deep learning and G2F publications hot off the press

P3 faculty are busy this summer publishing their research findings. In this edition of faculty publication news, three new publications are out from three teams of researchers. 

 

P3 Co-PIs Carolyn Lawrence-Dill and Patrick Schnable published their scientific perspective on ways to increase agricultural productivity in an environmentally sustainable fashion; namely with their Genomes to Fields Initiative. The maize Genomes to Fields (G2F) Initiative seeks to bring together researchers across diverse disciplines to create situations that generate new ideas for research, development, and education.

 

In the journal PLANT PHENOMICS, P3 faculty affiliates Danny Singh, Arti Singh, Baskar Ganapathysubramanian and Soumik Sarkar published their collaborative, interdisciplinary work on "A Weakly Supervised Deep Learning Framework for Sorghum Head Detection and Counting". In this work, the researchers, with their other team members, developed a network that can measure certain phenotypic traits at a greater accuracy than what can be done by humans. Counting cereal crop heads manually is an extremely labor-intensive process and suffers from low efficiency and human errors. Their model is able to learn what constitutes a crop head without relying on as many hand-labeled images during training as what has been previously required.

 

P3 faculty affiliate, Steven Cannon, published his work on "Genome-wide associations and epistatic interactions for internode number, plant height, seed weight and seed yield in soybean" in BMC Genomics. In this paper, with his team, Cannon describes the utility of GWAS and GWES approaches for identifying marker-trait associations and interactions within a diverse germplasm collection.