Ensuring High Quality in Masa

Humans directly consume corn through many different food products such as tortillas, corn chips, tamales, and more that require high quality, high protein grain. However, breeding companies often do not consider grain composition traits when developing elite germplasm, instead opting to increase yield, which can negatively impact certain quality components such as protein content. This is particularly damaging to masa-based products, a significant source of calories, calcium, and niacin for many parts of the world. Masa-based products must maintain quality standards such as moisture content and pericarp (outer layer of the seed) retention due to their impact on other traits such as taste, texture, and nutrition. Manufacturers can account for minor deficiencies in composition by changing cooking parameters during nixtamalization, a high-heat, high-pH cooking process used to create masa. However, as breeding companies continue to breed for higher yields, nixtamalization parameter tuning becomes more difficult and will eventually be inadequate. If breeding companies are to consider masa quality traits during germplasm development, tools to analyze a small quantity of a large number of samples are required

Informatics-based solutions such as machine learning and image analysis are being developed that will allow users to predict how much moisture will be absorbed during the cooking process and assess how easily the pericarp is lost during the nixtamalization process. PhD student Michael Burns (Applied Plant Sciences), in a project called “Developing Tools to Predictively Assess Masa Quality Through Machine Learning and Computer Visions,” is working to create methods that will allow breeders to make high selection intensity decisions for both masa quality and yield which will benefit growers, manufacturers, and consumers around the world.

Some funding for this project was provided by a 2023 University of Minnesota Informatics Institute MnDRIVE PhD Graduate Assistantship. The UMII MnDRIVE Graduate Assistantship program supports U of M PhD candidates pursuing research at the intersection of informatics and any of the five MnDRIVE areas:

• Robotics
• Global Food
• Environment
• Brain Conditions
• Cancer Clinical Trials

This project is part of the Global Food MnDRIVE area.

See the complete list of the UMII Graduate Assistantships for 2023.

Image description: Example of image analysis pipeline for a single sample. An image of 50 kernels is taken at 1200 dpi, the white background is removed from the image, and each kernel is separated into its own image. Manual tracing of the stained pericarp is performed followed by the removal of pixels background pixels from the polygon. From this polygon, a pixel-wise mask can be created to differentiate pericarp from non-pericarp pixels.