Latest paper from our group: Computer vision and machine learning for robust phenotyping in genome-wide studies
Latest paper from our group was published in Scientific Reports. This work is an example of leveraging cross-discipline expertise to tackle big problems and also for the development of phenomics capabilities.
In this paper, we show the development of a machine learning (ML)-enabled image-phenotyping pipeline for soybean genetic studies. ML-generated phenotypic data were utilized for the genome-wide association study and genomic prediction, and we illustrated the reliability and advantage of ML-enabled image-phenotyping pipeline by identifying previously reported locus and a novel locus for an abiotic stress trait in soybean. This study demonstrates a promising path for advancing the field of phenomics and its integration with genomic prediction by providing an example of a systematic framework enabling robust and quicker phenotyping through ground- and air-based systems.
First author Dr. J. Zhang is a PDF in our group, while two other PDFs from our group Dr. Assefa (currently at USDA-ARS) and Dr. Chowda Reddy are co-authors. Collaborators and co-corresponding authors include Dr. Arti Singh (Agronomy) and Dr. B. Ganapathysubramanian (ME). Other authors include Mr. HS Naik, graduated with MS degree from the ME program at ISU, and Dr. S. Sarkar is a faculty member in college of engineering.
Congratulation to all authors for their contributions, and special thanks to the funding agencies for enabling this inter-disciplinary work!!
Full paper can be read at: http://www.nature.com/articles/srep44048