Michael Burns, Ph.D. Candidate at the University of Minnesota.

Each month, NCI updates the “Cereal Innovators” series that offers insight into the world of baking and cereal grains. During the webinar, participants were instructed on the topic of “Understanding Kernel Properties Affecting Masa Based Products.” This week’s presentation featured Michael Burns, Ph.D. Candidate at the University of Minnesota.

Burns began by providing background on what masa flour is used in along with the composition of an individual kernel of corn. Masa flour is used most typically in corn chips (though it’s found in other products), and the composition of a kernel of corn always includes the same parts. These parts are as follows: the pericarp (the outermost tissue of the kernel), the endosperm (both hard and soft), and the germ (the part of the kernel that develops into the new corn plant). Burns also provided a breakdown of what materials compose each part of the kernel.

From there, he discussed the process of Nixtamalization, which is the cooking process used to prepare kernels for various foods. During this process, the pericarp is removed, making the kernel softer and more easily useable in food products. Once the kernels are cooked, they must pass a number of tests in order to be deemed suitable for consumption. These include stress cracking, kernel shape, moisture content, pericarp retention, test weight, and others.

For the purposes of this webinar, Burns largely focused on the moisture content test, which he called the “goldilocks” test. This is because the moisture content of the kernel needs to be in the middle zone of the spectrum – not too high, and not too low, and around 45-50% moisture, depending on the company and the product being made. Too much or too little moisture can affect the physical appearance of the product along with its taste and other properties.

Moisture content can be controlled in a variety of ways, such as controlling how long the kernels are cooked for. Giving the product more time to cook and steep allows for more moisture to make its way into the product. Adjusting the temperature that the kernels are cooked at also has an effect on the moisture content of the kernels, as can how much lime solution is added to the mixture (this affects the pH level of the solution, which makes the pericarp dissolve faster and allows more moisture inside).

Burns added that food-grade corn kernels rarely see breeding work. This is because of the types of uses that food-grade corn is used for. When breeding work is done, though, the process involves the manufacturer sending samples of corn bred for various uses to pilot plant trials to determine which works best for each possible use of corn. These samples are then cooked and tested for the properties listed above to determine their quality and potential for use in that specific purpose. The only downside – and the reason that this isn’t done frequently – is that it’s a time-consuming process that also requires significant sample sizes, both of which limit its appeal.

This is the exact focus of Burns’ work. This work focuses on reducing the two issues with corn breeding work highlighted previously: the large sample size and lengthy time requirements. Instead of requiring 500 pounds of corn (the current requirement for most pilot test trials), he aims to lower that requirement down to 100 grams of corn – a significant reduction. Using these smaller sample sizes makes it possible to reduce the time required for each test trial, making it feasible to test hundreds of samples per day rather than one hundred samples per season.

This begs the question: how do we accomplish this? How do we shorten the time needed for these test trials while also reducing the sample size required for adequate results? Burns explains that, since moisture content in corn kernels is dependent on their composition, a technique called near-infrared (NIR) spectroscopy can be used to measure the type and quantities of atomic bonds in the kernels. This, in theory, should be able to help scientists predict a composition based-trait of the kernels, such as moisture content. And, at the same time, allow for any possible errors that are generated during the test to be minimized or perhaps even eliminated.

Process of using NIR Spectroscopy and classical machine learning to predict corn kernel traits.

There’s something else that, in this scenario, needs to be used in conjunction with NIR spectroscopy in order for the test to work: classical machine learning. This is used to predict the traits of the samples based on the properties observed using the NIR spectroscopy. Using this pairing to accurately predict the moisture content of corn kernels also makes it possible to answer what Burns referred to as “deeper biological questions.”

These questions include ones such as, “What explains more variation: genetics or the environment?” Burns’ answer to this states that the environment appears to explain most of the moisture content variation, not genetics. Another question this process was able to answer is, “If moisture content is so heavily controlled by the environment, what should we do?” His answer includes three possible strategies: 1) ignore it (to improve average performance and focus on changing the cooking parameters); 2) control it (to breed for stability in an environmental response); or 3) exploit it (to source corn from where the environment is advantageous and to breed there as well).

Stages of the food-grade production process.

While this process of machine learning and NIR spectroscopy is very promising, Burns stated that it still needs to be validated on hybrid corn species, something that he intends to pursue in the future. He also reiterated the current promise that this method has for non-hybrid corn species, and that his research can be implemented at numerous stages of the food-grade corn production process. This demands further research and testing into this method for testing corn kernels for moisture content.

The Northern Crops Institute greatly appreciates Michael Burns’ professional input and involvement in our webinar series. At NCI, we continue to fulfill our mission to support regional agriculture and value-added processing by conducting educational and technical programs that expand and maintain domestic and international markets for northern grown crops.

For more information about future webinars offered at NCI, click here.

To view yesterday’s webinar, click the recording below.

NCI Cereal Innovators Webinar Series | Featuring Michael Burns

© 2024 NORTHERN CROPS INSTITUTE