Using Machine Learning methods in our research at CiBreed

Machine Learning (ML), a buzz word one most likely came across a few times in the last year. Academic and industry professionals in various fields have published numerous articles illustrating how ML can address various prediction problems or extract insightful patterns from complex datasets. In this article, we discuss ML and its applications to plant and animal breeding.

What is ML?

ML can be defined as the study of algorithms that inherently enhance their performance through experience obtained from data (Mitchell, 1997), and attempt at identifying patterns or learn complex relationships that are not explicitly stated in the model. Given the diversity and the large amount of data generated in animal and plant breeding today, machine learning has the potential to serve diverse needs ranging from automated phenotyping, modeling complex interactions such as epistasis or dominance, or analyzing data of mixed types .

To understand how ML can help us understand data generated from breeding, we had a discussion with CiBreed researchers who implement ML algorithms in their work.

Authors: Cathy Westhues, Damilola Adekale and Selina Klees. Download full article about “Machine Learning”